v0.23.0

Added:
- Inspector.DeleteAllAggregatingTasks
- Inspector.ArchiveAllAggregatingTasks
- Inspector.RunAllAggregatingTasks

.github/workflows/build.yml

@@ -7,7 +7,7 @@ jobs:
     strategy:
       matrix:
         os: [ubuntu-latest]
-        go-version: [1.14.x, 1.15.x, 1.16.x, 1.17.x]
+        go-version: [1.14.x, 1.15.x, 1.16.x, 1.17.x, 1.18.x]
     runs-on: ${{ matrix.os }}
     services:
       redis:

CHANGELOG.md

@@ -7,6 +7,23 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.23.0] - 2022-03-11
+
+### Added
+
+- `Group` option is introduced to enqueue task in a group.
+- `GroupAggregator` and related types are introduced for task aggregation feature.
+- `GroupGracePeriod`, `GroupMaxSize`, `GroupMaxDelay`, and `GroupAggregator` fields are added to `Config`.
+- `Inspector` has new methods related to "aggregating tasks".
+- `Group` field is added to `TaskInfo`.
+- (CLI): `group ls` command is added
+- (CLI): `task ls` supports listing aggregating tasks via `--state=aggregating --group=<GROUP>` flags
+- Enable rediss url parsing support
+
+### Fixed
+
+- Fixed overflow issue with 32-bit systems (For details, see https://github.com/hibiken/asynq/pull/426)
+
 ## [0.22.1] - 2022-02-20
 
 ### Fixed
@@ -264,9 +281,9 @@ Use `ProcessIn` or `ProcessAt` option to schedule a task instead of `EnqueueIn`
 
 #### `Inspector`
 
-All Inspector methods are scoped to a queue, and the methods take `qname (string)` as the first argument.  
-`EnqueuedTask` is renamed to `PendingTask` and its corresponding methods.  
-`InProgressTask` is renamed to `ActiveTask` and its corresponding methods.  
+All Inspector methods are scoped to a queue, and the methods take `qname (string)` as the first argument.
+`EnqueuedTask` is renamed to `PendingTask` and its corresponding methods.
+`InProgressTask` is renamed to `ActiveTask` and its corresponding methods.
 Command "Enqueue" is replaced by the verb "Run" (e.g. `EnqueueAllScheduledTasks` --> `RunAllScheduledTasks`)
 
 #### `CLI`

aggregator.go

@@ -0,0 +1,176 @@
+// Copyright 2022 Kentaro Hibino. All rights reserved.
+// Use of this source code is governed by a MIT license
+// that can be found in the LICENSE file.
+
+package asynq
+
+import (
+	"context"
+	"sync"
+	"time"
+
+	"github.com/hibiken/asynq/internal/base"
+	"github.com/hibiken/asynq/internal/log"
+)
+
+// An aggregator is responsible for checking groups and aggregate into one task
+// if any of the grouping condition is met.
+type aggregator struct {
+	logger *log.Logger
+	broker base.Broker
+	client *Client
+
+	// channel to communicate back to the long running "aggregator" goroutine.
+	done chan struct{}
+
+	// list of queue names to check and aggregate.
+	queues []string
+
+	// Group configurations
+	gracePeriod time.Duration
+	maxDelay    time.Duration
+	maxSize     int
+
+	// User provided group aggregator.
+	ga GroupAggregator
+
+	// interval used to check for aggregation
+	interval time.Duration
+
+	// sema is a counting semaphore to ensure the number of active aggregating function
+	// does not exceed the limit.
+	sema chan struct{}
+}
+
+type aggregatorParams struct {
+	logger          *log.Logger
+	broker          base.Broker
+	queues          []string
+	gracePeriod     time.Duration
+	maxDelay        time.Duration
+	maxSize         int
+	groupAggregator GroupAggregator
+}
+
+const (
+	// Maximum number of aggregation checks in flight concurrently.
+	maxConcurrentAggregationChecks = 3
+
+	// Default interval used for aggregation checks. If the provided gracePeriod is less than
+	// the default, use the gracePeriod.
+	defaultAggregationCheckInterval = 7 * time.Second
+)
+
+func newAggregator(params aggregatorParams) *aggregator {
+	interval := defaultAggregationCheckInterval
+	if params.gracePeriod < interval {
+		interval = params.gracePeriod
+	}
+	return &aggregator{
+		logger:      params.logger,
+		broker:      params.broker,
+		client:      &Client{broker: params.broker},
+		done:        make(chan struct{}),
+		queues:      params.queues,
+		gracePeriod: params.gracePeriod,
+		maxDelay:    params.maxDelay,
+		maxSize:     params.maxSize,
+		ga:          params.groupAggregator,
+		sema:        make(chan struct{}, maxConcurrentAggregationChecks),
+		interval:    interval,
+	}
+}
+
+func (a *aggregator) shutdown() {
+	if a.ga == nil {
+		return
+	}
+	a.logger.Debug("Aggregator shutting down...")
+	// Signal the aggregator goroutine to stop.
+	a.done <- struct{}{}
+}
+
+func (a *aggregator) start(wg *sync.WaitGroup) {
+	if a.ga == nil {
+		return
+	}
+	wg.Add(1)
+	go func() {
+		defer wg.Done()
+		ticker := time.NewTicker(a.interval)
+		for {
+			select {
+			case <-a.done:
+				a.logger.Debug("Waiting for all aggregation checks to finish...")
+				// block until all aggregation checks released the token
+				for i := 0; i < cap(a.sema); i++ {
+					a.sema <- struct{}{}
+				}
+				a.logger.Debug("Aggregator done")
+				ticker.Stop()
+				return
+			case t := <-ticker.C:
+				a.exec(t)
+			}
+		}
+	}()
+}
+
+func (a *aggregator) exec(t time.Time) {
+	select {
+	case a.sema <- struct{}{}: // acquire token
+		go a.aggregate(t)
+	default:
+		// If the semaphore blocks, then we are currently running max number of
+		// aggregation checks. Skip this round and log warning.
+		a.logger.Warnf("Max number of aggregation checks in flight. Skipping")
+	}
+}
+
+func (a *aggregator) aggregate(t time.Time) {
+	defer func() { <-a.sema /* release token */ }()
+	for _, qname := range a.queues {
+		groups, err := a.broker.ListGroups(qname)
+		if err != nil {
+			a.logger.Errorf("Failed to list groups in queue: %q", qname)
+			continue
+		}
+		for _, gname := range groups {
+			aggregationSetID, err := a.broker.AggregationCheck(
+				qname, gname, t, a.gracePeriod, a.maxDelay, a.maxSize)
+			if err != nil {
+				a.logger.Errorf("Failed to run aggregation check: queue=%q group=%q", qname, gname)
+				continue
+			}
+			if aggregationSetID == "" {
+				a.logger.Debugf("No aggregation needed at this time: queue=%q group=%q", qname, gname)
+				continue
+			}
+
+			// Aggregate and enqueue.
+			msgs, deadline, err := a.broker.ReadAggregationSet(qname, gname, aggregationSetID)
+			if err != nil {
+				a.logger.Errorf("Failed to read aggregation set: queue=%q, group=%q, setID=%q",
+					qname, gname, aggregationSetID)
+				continue
+			}
+			tasks := make([]*Task, len(msgs))
+			for i, m := range msgs {
+				tasks[i] = NewTask(m.Type, m.Payload)
+			}
+			aggregatedTask := a.ga.Aggregate(gname, tasks)
+			ctx, cancel := context.WithDeadline(context.Background(), deadline)
+			if _, err := a.client.EnqueueContext(ctx, aggregatedTask, Queue(qname)); err != nil {
+				a.logger.Errorf("Failed to enqueue aggregated task (queue=%q, group=%q, setID=%q): %v",
+					qname, gname, aggregationSetID, err)
+				cancel()
+				continue
+			}
+			if err := a.broker.DeleteAggregationSet(ctx, qname, gname, aggregationSetID); err != nil {
+				a.logger.Warnf("Failed to delete aggregation set: queue=%q, group=%q, setID=%q",
+					qname, gname, aggregationSetID)
+			}
+			cancel()
+		}
+	}
+}

aggregator_test.go

@@ -0,0 +1,165 @@
+// Copyright 2022 Kentaro Hibino. All rights reserved.
+// Use of this source code is governed by a MIT license
+// that can be found in the LICENSE file.
+
+package asynq
+
+import (
+	"sync"
+	"testing"
+	"time"
+
+	"github.com/google/go-cmp/cmp"
+	"github.com/hibiken/asynq/internal/base"
+	"github.com/hibiken/asynq/internal/rdb"
+	h "github.com/hibiken/asynq/internal/testutil"
+)
+
+func TestAggregator(t *testing.T) {
+	r := setup(t)
+	defer r.Close()
+	rdbClient := rdb.NewRDB(r)
+	client := Client{broker: rdbClient}
+
+	tests := []struct {
+		desc             string
+		gracePeriod      time.Duration
+		maxDelay         time.Duration
+		maxSize          int
+		aggregateFunc    func(gname string, tasks []*Task) *Task
+		tasks            []*Task       // tasks to enqueue
+		enqueueFrequency time.Duration // time between one enqueue event to another
+		waitTime         time.Duration // time to wait
+		wantGroups       map[string]map[string][]base.Z
+		wantPending      map[string][]*base.TaskMessage
+	}{
+		{
+			desc:        "group older than the grace period should be aggregated",
+			gracePeriod: 1 * time.Second,
+			maxDelay:    0, // no maxdelay limit
+			maxSize:     0, // no maxsize limit
+			aggregateFunc: func(gname string, tasks []*Task) *Task {
+				return NewTask(gname, nil, MaxRetry(len(tasks))) // use max retry to see how many tasks were aggregated
+			},
+			tasks: []*Task{
+				NewTask("task1", nil, Group("mygroup")),
+				NewTask("task2", nil, Group("mygroup")),
+				NewTask("task3", nil, Group("mygroup")),
+			},
+			enqueueFrequency: 300 * time.Millisecond,
+			waitTime:         3 * time.Second,
+			wantGroups: map[string]map[string][]base.Z{
+				"default": {
+					"mygroup": {},
+				},
+			},
+			wantPending: map[string][]*base.TaskMessage{
+				"default": {
+					h.NewTaskMessageBuilder().SetType("mygroup").SetRetry(3).Build(),
+				},
+			},
+		},
+		{
+			desc:        "group older than the max-delay should be aggregated",
+			gracePeriod: 2 * time.Second,
+			maxDelay:    4 * time.Second,
+			maxSize:     0, // no maxsize limit
+			aggregateFunc: func(gname string, tasks []*Task) *Task {
+				return NewTask(gname, nil, MaxRetry(len(tasks))) // use max retry to see how many tasks were aggregated
+			},
+			tasks: []*Task{
+				NewTask("task1", nil, Group("mygroup")), // time 0
+				NewTask("task2", nil, Group("mygroup")), // time 1s
+				NewTask("task3", nil, Group("mygroup")), // time 2s
+				NewTask("task4", nil, Group("mygroup")), // time 3s
+			},
+			enqueueFrequency: 1 * time.Second,
+			waitTime:         4 * time.Second,
+			wantGroups: map[string]map[string][]base.Z{
+				"default": {
+					"mygroup": {},
+				},
+			},
+			wantPending: map[string][]*base.TaskMessage{
+				"default": {
+					h.NewTaskMessageBuilder().SetType("mygroup").SetRetry(4).Build(),
+				},
+			},
+		},
+		{
+			desc:        "group reached the max-size should be aggregated",
+			gracePeriod: 1 * time.Minute,
+			maxDelay:    0, // no maxdelay limit
+			maxSize:     5,
+			aggregateFunc: func(gname string, tasks []*Task) *Task {
+				return NewTask(gname, nil, MaxRetry(len(tasks))) // use max retry to see how many tasks were aggregated
+			},
+			tasks: []*Task{
+				NewTask("task1", nil, Group("mygroup")),
+				NewTask("task2", nil, Group("mygroup")),
+				NewTask("task3", nil, Group("mygroup")),
+				NewTask("task4", nil, Group("mygroup")),
+				NewTask("task5", nil, Group("mygroup")),
+			},
+			enqueueFrequency: 300 * time.Millisecond,
+			waitTime:         defaultAggregationCheckInterval * 2,
+			wantGroups: map[string]map[string][]base.Z{
+				"default": {
+					"mygroup": {},
+				},
+			},
+			wantPending: map[string][]*base.TaskMessage{
+				"default": {
+					h.NewTaskMessageBuilder().SetType("mygroup").SetRetry(5).Build(),
+				},
+			},
+		},
+	}
+
+	for _, tc := range tests {
+		h.FlushDB(t, r)
+
+		aggregator := newAggregator(aggregatorParams{
+			logger:          testLogger,
+			broker:          rdbClient,
+			queues:          []string{"default"},
+			gracePeriod:     tc.gracePeriod,
+			maxDelay:        tc.maxDelay,
+			maxSize:         tc.maxSize,
+			groupAggregator: GroupAggregatorFunc(tc.aggregateFunc),
+		})
+
+		var wg sync.WaitGroup
+		aggregator.start(&wg)
+
+		for _, task := range tc.tasks {
+			if _, err := client.Enqueue(task); err != nil {
+				t.Errorf("%s: Client Enqueue failed: %v", tc.desc, err)
+				aggregator.shutdown()
+				wg.Wait()
+				continue
+			}
+			time.Sleep(tc.enqueueFrequency)
+		}
+
+		time.Sleep(tc.waitTime)
+
+		for qname, groups := range tc.wantGroups {
+			for gname, want := range groups {
+				gotGroup := h.GetGroupEntries(t, r, qname, gname)
+				if diff := cmp.Diff(want, gotGroup, h.SortZSetEntryOpt); diff != "" {
+					t.Errorf("%s: mismatch found in %q; (-want,+got)\n%s", tc.desc, base.GroupKey(qname, gname), diff)
+				}
+			}
+		}
+
+		for qname, want := range tc.wantPending {
+			gotPending := h.GetPendingMessages(t, r, qname)
+			if diff := cmp.Diff(want, gotPending, h.SortMsgOpt, h.IgnoreIDOpt); diff != "" {
+				t.Errorf("%s: mismatch found in %q; (-want,+got)\n%s", tc.desc, base.PendingKey(qname), diff)
+			}
+		}
+		aggregator.shutdown()
+		wg.Wait()
+	}
+}

asynq.go

@@ -8,6 +8,7 @@ import (
 	"context"
 	"crypto/tls"
 	"fmt"
+	"net"
 	"net/url"
 	"strconv"
 	"strings"
@@ -97,6 +98,14 @@ type TaskInfo struct {
 	// Deadline is the deadline for the task, zero value if not specified.
 	Deadline time.Time
 
+	// Group is the name of the group in which the task belongs.
+	//
+	// Tasks in the same queue can be grouped together by Group name and will be aggregated into one task
+	// by a Server processing the queue.
+	//
+	// Empty string (default) indicates task does not belong to any groups, and no aggregation will be applied to the task.
+	Group string
+
 	// NextProcessAt is the time the task is scheduled to be processed,
 	// zero if not applicable.
 	NextProcessAt time.Time
@@ -139,6 +148,7 @@ func newTaskInfo(msg *base.TaskMessage, state base.TaskState, nextProcessAt time
 		MaxRetry:      msg.Retry,
 		Retried:       msg.Retried,
 		LastErr:       msg.ErrorMsg,
+		Group:         msg.GroupKey,
 		Timeout:       time.Duration(msg.Timeout) * time.Second,
 		Deadline:      fromUnixTimeOrZero(msg.Deadline),
 		Retention:     time.Duration(msg.Retention) * time.Second,
@@ -161,6 +171,8 @@ func newTaskInfo(msg *base.TaskMessage, state base.TaskState, nextProcessAt time
 		info.State = TaskStateArchived
 	case base.TaskStateCompleted:
 		info.State = TaskStateCompleted
+	case base.TaskStateAggregating:
+		info.State = TaskStateAggregating
 	default:
 		panic(fmt.Sprintf("internal error: unknown state: %d", state))
 	}
@@ -188,6 +200,9 @@ const (
 
 	// Indicates that the task is processed successfully and retained until the retention TTL expires.
 	TaskStateCompleted
+
+	// Indicates that the task is waiting in a group to be aggreated into one task.
+	TaskStateAggregating
 )
 
 func (s TaskState) String() string {
@@ -204,6 +219,8 @@ func (s TaskState) String() string {
 		return "archived"
 	case TaskStateCompleted:
 		return "completed"
+	case TaskStateAggregating:
+		return "aggregating"
 	}
 	panic("asynq: unknown task state")
 }
@@ -419,9 +436,10 @@ func (opt RedisClusterClientOpt) MakeRedisClient() interface{} {
 // ParseRedisURI parses redis uri string and returns RedisConnOpt if uri is valid.
 // It returns a non-nil error if uri cannot be parsed.
 //
-// Three URI schemes are supported, which are redis:, redis-socket:, and redis-sentinel:.
+// Three URI schemes are supported, which are redis:, rediss:, redis-socket:, and redis-sentinel:.
 // Supported formats are:
 //     redis://[:password@]host[:port][/dbnumber]
+//     rediss://[:password@]host[:port][/dbnumber]
 //     redis-socket://[:password@]path[?db=dbnumber]
 //     redis-sentinel://[:password@]host1[:port][,host2:[:port]][,hostN:[:port]][?master=masterName]
 func ParseRedisURI(uri string) (RedisConnOpt, error) {
@@ -430,7 +448,7 @@ func ParseRedisURI(uri string) (RedisConnOpt, error) {
 		return nil, fmt.Errorf("asynq: could not parse redis uri: %v", err)
 	}
 	switch u.Scheme {
-	case "redis":
+	case "redis", "rediss":
 		return parseRedisURI(u)
 	case "redis-socket":
 		return parseRedisSocketURI(u)
@@ -444,6 +462,8 @@ func ParseRedisURI(uri string) (RedisConnOpt, error) {
 func parseRedisURI(u *url.URL) (RedisConnOpt, error) {
 	var db int
 	var err error
+	var redisConnOpt RedisClientOpt
+
 	if len(u.Path) > 0 {
 		xs := strings.Split(strings.Trim(u.Path, "/"), "/")
 		db, err = strconv.Atoi(xs[0])
@@ -455,7 +475,20 @@ func parseRedisURI(u *url.URL) (RedisConnOpt, error) {
 	if v, ok := u.User.Password(); ok {
 		password = v
 	}
-	return RedisClientOpt{Addr: u.Host, DB: db, Password: password}, nil
+
+	if u.Scheme == "rediss" {
+		h, _, err := net.SplitHostPort(u.Host)
+		if err != nil {
+			h = u.Host
+		}
+		redisConnOpt.TLSConfig = &tls.Config{ServerName: h}
+	}
+
+	redisConnOpt.Addr = u.Host
+	redisConnOpt.Password = password
+	redisConnOpt.DB = db
+
+	return redisConnOpt, nil
 }
 
 func parseRedisSocketURI(u *url.URL) (RedisConnOpt, error) {

asynq_test.go

@@ -5,6 +5,7 @@
 package asynq
 
 import (
+	"crypto/tls"
 	"flag"
 	"sort"
 	"strings"
@@ -12,8 +13,9 @@ import (
 
 	"github.com/go-redis/redis/v8"
 	"github.com/google/go-cmp/cmp"
-	h "github.com/hibiken/asynq/internal/asynqtest"
+	"github.com/google/go-cmp/cmp/cmpopts"
 	"github.com/hibiken/asynq/internal/log"
+	h "github.com/hibiken/asynq/internal/testutil"
 )
 
 //============================================================================
@@ -99,6 +101,10 @@ func TestParseRedisURI(t *testing.T) {
 			"redis://localhost:6379",
 			RedisClientOpt{Addr: "localhost:6379"},
 		},
+		{
+			"rediss://localhost:6379",
+			RedisClientOpt{Addr: "localhost:6379", TLSConfig: &tls.Config{ServerName: "localhost"}},
+		},
 		{
 			"redis://localhost:6379/3",
 			RedisClientOpt{Addr: "localhost:6379", DB: 3},
@@ -151,7 +157,7 @@ func TestParseRedisURI(t *testing.T) {
 			continue
 		}
 
-		if diff := cmp.Diff(tc.want, got); diff != "" {
+		if diff := cmp.Diff(tc.want, got, cmpopts.IgnoreUnexported(tls.Config{})); diff != "" {
 			t.Errorf("ParseRedisURI(%q) = %+v, want %+v\n(-want,+got)\n%s", tc.uri, got, tc.want, diff)
 		}
 	}

benchmark_test.go

@@ -12,7 +12,7 @@ import (
 	"testing"
 	"time"
 
-	h "github.com/hibiken/asynq/internal/asynqtest"
+	h "github.com/hibiken/asynq/internal/testutil"
 )
 
 // Creates a new task of type "task<n>" with payload {"data": n}.

client.go

@@ -24,7 +24,7 @@ import (
 //
 // Clients are safe for concurrent use by multiple goroutines.
 type Client struct {
-	rdb *rdb.RDB
+	broker base.Broker
 }
 
 // NewClient returns a new Client instance given a redis connection option.
@@ -33,7 +33,7 @@ func NewClient(r RedisConnOpt) *Client {
 	if !ok {
 		panic(fmt.Sprintf("asynq: unsupported RedisConnOpt type %T", r))
 	}
-	return &Client{rdb: rdb.NewRDB(c)}
+	return &Client{broker: rdb.NewRDB(c)}
 }
 
 type OptionType int
@@ -48,6 +48,7 @@ const (
 	ProcessInOpt
 	TaskIDOpt
 	RetentionOpt
+	GroupOpt
 )
 
 // Option specifies the task processing behavior.
@@ -73,6 +74,7 @@ type (
 	processAtOption time.Time
 	processInOption time.Duration
 	retentionOption time.Duration
+	groupOption     string
 )
 
 // MaxRetry returns an option to specify the max number of times
@@ -91,13 +93,13 @@ func (n retryOption) Type() OptionType   { return MaxRetryOpt }
 func (n retryOption) Value() interface{} { return int(n) }
 
 // Queue returns an option to specify the queue to enqueue the task into.
-func Queue(qname string) Option {
-	return queueOption(qname)
+func Queue(name string) Option {
+	return queueOption(name)
 }
 
-func (qname queueOption) String() string     { return fmt.Sprintf("Queue(%q)", string(qname)) }
-func (qname queueOption) Type() OptionType   { return QueueOpt }
-func (qname queueOption) Value() interface{} { return string(qname) }
+func (name queueOption) String() string     { return fmt.Sprintf("Queue(%q)", string(name)) }
+func (name queueOption) Type() OptionType   { return QueueOpt }
+func (name queueOption) Value() interface{} { return string(name) }
 
 // TaskID returns an option to specify the task ID.
 func TaskID(id string) Option {
@@ -142,7 +144,8 @@ func (t deadlineOption) Value() interface{} { return time.Time(t) }
 
 // Unique returns an option to enqueue a task only if the given task is unique.
 // Task enqueued with this option is guaranteed to be unique within the given ttl.
-// Once the task gets processed successfully or once the TTL has expired, another task with the same uniqueness may be enqueued.
+// Once the task gets processed successfully or once the TTL has expired,
+// another task with the same uniqueness may be enqueued.
 // ErrDuplicateTask error is returned when enqueueing a duplicate task.
 // TTL duration must be greater than or equal to 1 second.
 //
@@ -193,6 +196,16 @@ func (ttl retentionOption) String() string     { return fmt.Sprintf("Retention(%
 func (ttl retentionOption) Type() OptionType   { return RetentionOpt }
 func (ttl retentionOption) Value() interface{} { return time.Duration(ttl) }
 
+// Group returns an option to specify the group used for the task.
+// Tasks in a given queue with the same group will be aggregated into one task before passed to Handler.
+func Group(name string) Option {
+	return groupOption(name)
+}
+
+func (name groupOption) String() string     { return fmt.Sprintf("Group(%q)", string(name)) }
+func (name groupOption) Type() OptionType   { return GroupOpt }
+func (name groupOption) Value() interface{} { return string(name) }
+
 // ErrDuplicateTask indicates that the given task could not be enqueued since it's a duplicate of another task.
 //
 // ErrDuplicateTask error only applies to tasks enqueued with a Unique option.
@@ -212,6 +225,7 @@ type option struct {
 	uniqueTTL time.Duration
 	processAt time.Time
 	retention time.Duration
+	group     string
 }
 
 // composeOptions merges user provided options into the default options
@@ -239,8 +253,8 @@ func composeOptions(opts ...Option) (option, error) {
 			res.queue = qname
 		case taskIDOption:
 			id := string(opt)
-			if err := validateTaskID(id); err != nil {
-				return option{}, err
+			if isBlank(id) {
+				return option{}, errors.New("task ID cannot be empty")
 			}
 			res.taskID = id
 		case timeoutOption:
@@ -259,6 +273,12 @@ func composeOptions(opts ...Option) (option, error) {
 			res.processAt = time.Now().Add(time.Duration(opt))
 		case retentionOption:
 			res.retention = time.Duration(opt)
+		case groupOption:
+			key := string(opt)
+			if isBlank(key) {
+				return option{}, errors.New("group key cannot be empty")
+			}
+			res.group = key
 		default:
 			// ignore unexpected option
 		}
@@ -266,12 +286,9 @@ func composeOptions(opts ...Option) (option, error) {
 	return res, nil
 }
 
-// validates user provided task ID string.
-func validateTaskID(id string) error {
-	if strings.TrimSpace(id) == "" {
-		return errors.New("task ID cannot be empty")
-	}
-	return nil
+// isBlank returns true if the given s is empty or consist of all whitespaces.
+func isBlank(s string) bool {
+	return strings.TrimSpace(s) == ""
 }
 
 const (
@@ -290,7 +307,7 @@ var (
 
 // Close closes the connection with redis.
 func (c *Client) Close() error {
-	return c.rdb.Close()
+	return c.broker.Close()
 }
 
 // Enqueue enqueues the given task to a queue.
@@ -322,6 +339,9 @@ func (c *Client) Enqueue(task *Task, opts ...Option) (*TaskInfo, error) {
 //
 // The first argument context applies to the enqueue operation. To specify task timeout and deadline, use Timeout and Deadline option instead.
 func (c *Client) EnqueueContext(ctx context.Context, task *Task, opts ...Option) (*TaskInfo, error) {
+	if task == nil {
+		return nil, fmt.Errorf("task cannot be nil")
+	}
 	if strings.TrimSpace(task.Type()) == "" {
 		return nil, fmt.Errorf("task typename cannot be empty")
 	}
@@ -356,17 +376,23 @@ func (c *Client) EnqueueContext(ctx context.Context, task *Task, opts ...Option)
 		Deadline:  deadline.Unix(),
 		Timeout:   int64(timeout.Seconds()),
 		UniqueKey: uniqueKey,
+		GroupKey:  opt.group,
 		Retention: int64(opt.retention.Seconds()),
 	}
 	now := time.Now()
 	var state base.TaskState
-	if opt.processAt.Before(now) || opt.processAt.Equal(now) {
+	if opt.processAt.After(now) {
+		err = c.schedule(ctx, msg, opt.processAt, opt.uniqueTTL)
+		state = base.TaskStateScheduled
+	} else if opt.group != "" {
+		// Use zero value for processAt since we don't know when the task will be aggregated and processed.
+		opt.processAt = time.Time{}
+		err = c.addToGroup(ctx, msg, opt.group, opt.uniqueTTL)
+		state = base.TaskStateAggregating
+	} else {
 		opt.processAt = now
 		err = c.enqueue(ctx, msg, opt.uniqueTTL)
 		state = base.TaskStatePending
-	} else {
-		err = c.schedule(ctx, msg, opt.processAt, opt.uniqueTTL)
-		state = base.TaskStateScheduled
 	}
 	switch {
 	case errors.Is(err, errors.ErrDuplicateTask):
@@ -381,15 +407,22 @@ func (c *Client) EnqueueContext(ctx context.Context, task *Task, opts ...Option)
 
 func (c *Client) enqueue(ctx context.Context, msg *base.TaskMessage, uniqueTTL time.Duration) error {
 	if uniqueTTL > 0 {
-		return c.rdb.EnqueueUnique(ctx, msg, uniqueTTL)
+		return c.broker.EnqueueUnique(ctx, msg, uniqueTTL)
 	}
-	return c.rdb.Enqueue(ctx, msg)
+	return c.broker.Enqueue(ctx, msg)
 }
 
 func (c *Client) schedule(ctx context.Context, msg *base.TaskMessage, t time.Time, uniqueTTL time.Duration) error {
 	if uniqueTTL > 0 {
 		ttl := t.Add(uniqueTTL).Sub(time.Now())
-		return c.rdb.ScheduleUnique(ctx, msg, t, ttl)
+		return c.broker.ScheduleUnique(ctx, msg, t, ttl)
 	}
-	return c.rdb.Schedule(ctx, msg, t)
+	return c.broker.Schedule(ctx, msg, t)
+}
+
+func (c *Client) addToGroup(ctx context.Context, msg *base.TaskMessage, group string, uniqueTTL time.Duration) error {
+	if uniqueTTL > 0 {
+		return c.broker.AddToGroupUnique(ctx, msg, group, uniqueTTL)
+	}
+	return c.broker.AddToGroup(ctx, msg, group)
 }

client_test.go

@@ -12,8 +12,8 @@ import (
 
 	"github.com/google/go-cmp/cmp"
 	"github.com/google/go-cmp/cmp/cmpopts"
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
+	h "github.com/hibiken/asynq/internal/testutil"
 )
 
 func TestClientEnqueueWithProcessAtOption(t *testing.T) {
@@ -478,6 +478,158 @@ func TestClientEnqueue(t *testing.T) {
 	}
 }
 
+func TestClientEnqueueWithGroupOption(t *testing.T) {
+	r := setup(t)
+	client := NewClient(getRedisConnOpt(t))
+	defer client.Close()
+
+	task := NewTask("mytask", []byte("foo"))
+	now := time.Now()
+
+	tests := []struct {
+		desc          string
+		task          *Task
+		opts          []Option
+		wantInfo      *TaskInfo
+		wantPending   map[string][]*base.TaskMessage
+		wantGroups    map[string]map[string][]base.Z // map queue name to a set of groups
+		wantScheduled map[string][]base.Z
+	}{
+		{
+			desc: "With only Group option",
+			task: task,
+			opts: []Option{
+				Group("mygroup"),
+			},
+			wantInfo: &TaskInfo{
+				Queue:         "default",
+				Group:         "mygroup",
+				Type:          task.Type(),
+				Payload:       task.Payload(),
+				State:         TaskStateAggregating,
+				MaxRetry:      defaultMaxRetry,
+				Retried:       0,
+				LastErr:       "",
+				LastFailedAt:  time.Time{},
+				Timeout:       defaultTimeout,
+				Deadline:      time.Time{},
+				NextProcessAt: time.Time{},
+			},
+			wantPending: map[string][]*base.TaskMessage{
+				"default": {}, // should not be pending
+			},
+			wantGroups: map[string]map[string][]base.Z{
+				"default": {
+					"mygroup": {
+						{
+							Message: &base.TaskMessage{
+								Type:     task.Type(),
+								Payload:  task.Payload(),
+								Retry:    defaultMaxRetry,
+								Queue:    "default",
+								Timeout:  int64(defaultTimeout.Seconds()),
+								Deadline: noDeadline.Unix(),
+								GroupKey: "mygroup",
+							},
+							Score: now.Unix(),
+						},
+					},
+				},
+			},
+			wantScheduled: map[string][]base.Z{
+				"default": {},
+			},
+		},
+		{
+			desc: "With Group and ProcessIn options",
+			task: task,
+			opts: []Option{
+				Group("mygroup"),
+				ProcessIn(30 * time.Minute),
+			},
+			wantInfo: &TaskInfo{
+				Queue:         "default",
+				Group:         "mygroup",
+				Type:          task.Type(),
+				Payload:       task.Payload(),
+				State:         TaskStateScheduled,
+				MaxRetry:      defaultMaxRetry,
+				Retried:       0,
+				LastErr:       "",
+				LastFailedAt:  time.Time{},
+				Timeout:       defaultTimeout,
+				Deadline:      time.Time{},
+				NextProcessAt: now.Add(30 * time.Minute),
+			},
+			wantPending: map[string][]*base.TaskMessage{
+				"default": {}, // should not be pending
+			},
+			wantGroups: map[string]map[string][]base.Z{
+				"default": {
+					"mygroup": {}, // should not be added to the group yet
+				},
+			},
+			wantScheduled: map[string][]base.Z{
+				"default": {
+					{
+						Message: &base.TaskMessage{
+							Type:     task.Type(),
+							Payload:  task.Payload(),
+							Retry:    defaultMaxRetry,
+							Queue:    "default",
+							Timeout:  int64(defaultTimeout.Seconds()),
+							Deadline: noDeadline.Unix(),
+							GroupKey: "mygroup",
+						},
+						Score: now.Add(30 * time.Minute).Unix(),
+					},
+				},
+			},
+		},
+	}
+
+	for _, tc := range tests {
+		h.FlushDB(t, r) // clean up db before each test case.
+
+		gotInfo, err := client.Enqueue(tc.task, tc.opts...)
+		if err != nil {
+			t.Error(err)
+			continue
+		}
+		cmpOptions := []cmp.Option{
+			cmpopts.IgnoreFields(TaskInfo{}, "ID"),
+			cmpopts.EquateApproxTime(500 * time.Millisecond),
+		}
+		if diff := cmp.Diff(tc.wantInfo, gotInfo, cmpOptions...); diff != "" {
+			t.Errorf("%s;\nEnqueue(task) returned %v, want %v; (-want,+got)\n%s",
+				tc.desc, gotInfo, tc.wantInfo, diff)
+		}
+
+		for qname, want := range tc.wantPending {
+			got := h.GetPendingMessages(t, r, qname)
+			if diff := cmp.Diff(want, got, h.IgnoreIDOpt, cmpopts.EquateEmpty()); diff != "" {
+				t.Errorf("%s;\nmismatch found in %q; (-want,+got)\n%s", tc.desc, base.PendingKey(qname), diff)
+			}
+		}
+
+		for qname, groups := range tc.wantGroups {
+			for groupKey, want := range groups {
+				got := h.GetGroupEntries(t, r, qname, groupKey)
+				if diff := cmp.Diff(want, got, h.IgnoreIDOpt, cmpopts.EquateEmpty()); diff != "" {
+					t.Errorf("%s;\nmismatch found in %q; (-want,+got)\n%s", tc.desc, base.GroupKey(qname, groupKey), diff)
+				}
+			}
+		}
+
+		for qname, want := range tc.wantScheduled {
+			gotScheduled := h.GetScheduledEntries(t, r, qname)
+			if diff := cmp.Diff(want, gotScheduled, h.IgnoreIDOpt, cmpopts.EquateEmpty()); diff != "" {
+				t.Errorf("%s;\nmismatch found in %q; (-want,+got)\n%s", tc.desc, base.ScheduledKey(qname), diff)
+			}
+		}
+	}
+}
+
 func TestClientEnqueueWithTaskIDOption(t *testing.T) {
 	r := setup(t)
 	client := NewClient(getRedisConnOpt(t))
@@ -707,6 +859,11 @@ func TestClientEnqueueError(t *testing.T) {
 		task *Task
 		opts []Option
 	}{
+		{
+			desc: "With nil task",
+			task: nil,
+			opts: []Option{},
+		},
 		{
 			desc: "With empty queue name",
 			task: task,

context.go

@@ -36,7 +36,7 @@ func GetMaxRetry(ctx context.Context) (n int, ok bool) {
 
 // GetQueueName extracts queue name from a context, if any.
 //
-// Return value qname indicates which queue the task was pulled from.
-func GetQueueName(ctx context.Context) (qname string, ok bool) {
+// Return value queue indicates which queue the task was pulled from.
+func GetQueueName(ctx context.Context) (queue string, ok bool) {
 	return asynqcontext.GetQueueName(ctx)
 }

forwarder.go

@@ -56,13 +56,15 @@ func (f *forwarder) start(wg *sync.WaitGroup) {
 	wg.Add(1)
 	go func() {
 		defer wg.Done()
+		timer := time.NewTimer(f.avgInterval)
 		for {
 			select {
 			case <-f.done:
 				f.logger.Debug("Forwarder done")
 				return
-			case <-time.After(f.avgInterval):
+			case <-timer.C:
 				f.exec()
+				timer.Reset(f.avgInterval)
 			}
 		}
 	}()

forwarder_test.go

@@ -10,9 +10,9 @@ import (
 	"time"
 
 	"github.com/google/go-cmp/cmp"
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/rdb"
+	h "github.com/hibiken/asynq/internal/testutil"
 )
 
 func TestForwarder(t *testing.T) {

heartbeat_test.go

@@ -12,10 +12,10 @@ import (
 
 	"github.com/google/go-cmp/cmp"
 	"github.com/google/go-cmp/cmp/cmpopts"
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/rdb"
 	"github.com/hibiken/asynq/internal/testbroker"
+	h "github.com/hibiken/asynq/internal/testutil"
 	"github.com/hibiken/asynq/internal/timeutil"
 )
 

inspector.go

@@ -43,7 +43,32 @@ func (i *Inspector) Queues() ([]string, error) {
 	return i.rdb.AllQueues()
 }
 
-// QueueInfo represents a state of queues at a certain time.
+// Groups returns a list of all groups within the given queue.
+func (i *Inspector) Groups(queue string) ([]*GroupInfo, error) {
+	stats, err := i.rdb.GroupStats(queue)
+	if err != nil {
+		return nil, err
+	}
+	var res []*GroupInfo
+	for _, s := range stats {
+		res = append(res, &GroupInfo{
+			Group: s.Group,
+			Size:  s.Size,
+		})
+	}
+	return res, nil
+}
+
+// GroupInfo represents a state of a group at a cerntain time.
+type GroupInfo struct {
+	// Name of the group.
+	Group string
+
+	// Size is the total number of tasks in the group.
+	Size int
+}
+
+// QueueInfo represents a state of a queue at a certain time.
 type QueueInfo struct {
 	// Name of the queue.
 	Queue string
@@ -56,9 +81,12 @@ type QueueInfo struct {
 	Latency time.Duration
 
 	// Size is the total number of tasks in the queue.
-	// The value is the sum of Pending, Active, Scheduled, Retry, and Archived.
+	// The value is the sum of Pending, Active, Scheduled, Retry, Aggregating and Archived.
 	Size int
 
+	// Groups is the total number of groups in the queue.
+	Groups int
+
 	// Number of pending tasks.
 	Pending int
 	// Number of active tasks.
@@ -71,6 +99,8 @@ type QueueInfo struct {
 	Archived int
 	// Number of stored completed tasks.
 	Completed int
+	// Number of aggregating tasks.
+	Aggregating int
 
 	// Total number of tasks being processed within the given date (counter resets daily).
 	// The number includes both succeeded and failed tasks.
@@ -92,11 +122,11 @@ type QueueInfo struct {
 }
 
 // GetQueueInfo returns current information of the given queue.
-func (i *Inspector) GetQueueInfo(qname string) (*QueueInfo, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) GetQueueInfo(queue string) (*QueueInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, err
 	}
-	stats, err := i.rdb.CurrentStats(qname)
+	stats, err := i.rdb.CurrentStats(queue)
 	if err != nil {
 		return nil, err
 	}
@@ -105,12 +135,14 @@ func (i *Inspector) GetQueueInfo(qname string) (*QueueInfo, error) {
 		MemoryUsage:    stats.MemoryUsage,
 		Latency:        stats.Latency,
 		Size:           stats.Size,
+		Groups:         stats.Groups,
 		Pending:        stats.Pending,
 		Active:         stats.Active,
 		Scheduled:      stats.Scheduled,
 		Retry:          stats.Retry,
 		Archived:       stats.Archived,
 		Completed:      stats.Completed,
+		Aggregating:    stats.Aggregating,
 		Processed:      stats.Processed,
 		Failed:         stats.Failed,
 		ProcessedTotal: stats.ProcessedTotal,
@@ -134,11 +166,11 @@ type DailyStats struct {
 }
 
 // History returns a list of stats from the last n days.
-func (i *Inspector) History(qname string, n int) ([]*DailyStats, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) History(queue string, n int) ([]*DailyStats, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, err
 	}
-	stats, err := i.rdb.HistoricalStats(qname, n)
+	stats, err := i.rdb.HistoricalStats(queue, n)
 	if err != nil {
 		return nil, err
 	}
@@ -175,13 +207,13 @@ var (
 // If the specified queue does not exist, DeleteQueue returns ErrQueueNotFound.
 // If force is set to false and the specified queue is not empty, DeleteQueue
 // returns ErrQueueNotEmpty.
-func (i *Inspector) DeleteQueue(qname string, force bool) error {
-	err := i.rdb.RemoveQueue(qname, force)
+func (i *Inspector) DeleteQueue(queue string, force bool) error {
+	err := i.rdb.RemoveQueue(queue, force)
 	if errors.IsQueueNotFound(err) {
-		return fmt.Errorf("%w: queue=%q", ErrQueueNotFound, qname)
+		return fmt.Errorf("%w: queue=%q", ErrQueueNotFound, queue)
 	}
 	if errors.IsQueueNotEmpty(err) {
-		return fmt.Errorf("%w: queue=%q", ErrQueueNotEmpty, qname)
+		return fmt.Errorf("%w: queue=%q", ErrQueueNotEmpty, queue)
 	}
 	return err
 }
@@ -190,8 +222,8 @@ func (i *Inspector) DeleteQueue(qname string, force bool) error {
 //
 // Returns an error wrapping ErrQueueNotFound if a queue with the given name doesn't exist.
 // Returns an error wrapping ErrTaskNotFound if a task with the given id doesn't exist in the queue.
-func (i *Inspector) GetTaskInfo(qname, id string) (*TaskInfo, error) {
-	info, err := i.rdb.GetTaskInfo(qname, id)
+func (i *Inspector) GetTaskInfo(queue, id string) (*TaskInfo, error) {
+	info, err := i.rdb.GetTaskInfo(queue, id)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -267,13 +299,13 @@ func Page(n int) ListOption {
 // ListPendingTasks retrieves pending tasks from the specified queue.
 //
 // By default, it retrieves the first 30 tasks.
-func (i *Inspector) ListPendingTasks(qname string, opts ...ListOption) ([]*TaskInfo, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ListPendingTasks(queue string, opts ...ListOption) ([]*TaskInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
 	opt := composeListOptions(opts...)
 	pgn := rdb.Pagination{Size: opt.pageSize, Page: opt.pageNum - 1}
-	infos, err := i.rdb.ListPending(qname, pgn)
+	infos, err := i.rdb.ListPending(queue, pgn)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -295,20 +327,20 @@ func (i *Inspector) ListPendingTasks(qname string, opts ...ListOption) ([]*TaskI
 // ListActiveTasks retrieves active tasks from the specified queue.
 //
 // By default, it retrieves the first 30 tasks.
-func (i *Inspector) ListActiveTasks(qname string, opts ...ListOption) ([]*TaskInfo, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ListActiveTasks(queue string, opts ...ListOption) ([]*TaskInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
 	opt := composeListOptions(opts...)
 	pgn := rdb.Pagination{Size: opt.pageSize, Page: opt.pageNum - 1}
-	infos, err := i.rdb.ListActive(qname, pgn)
+	infos, err := i.rdb.ListActive(queue, pgn)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
 	case err != nil:
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
-	expired, err := i.rdb.ListLeaseExpired(time.Now(), qname)
+	expired, err := i.rdb.ListLeaseExpired(time.Now(), queue)
 	if err != nil {
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
@@ -332,17 +364,45 @@ func (i *Inspector) ListActiveTasks(qname string, opts ...ListOption) ([]*TaskIn
 	return tasks, nil
 }
 
-// ListScheduledTasks retrieves scheduled tasks from the specified queue.
-// Tasks are sorted by NextProcessAt in ascending order.
+// ListAggregatingTasks retrieves scheduled tasks from the specified group.
 //
 // By default, it retrieves the first 30 tasks.
-func (i *Inspector) ListScheduledTasks(qname string, opts ...ListOption) ([]*TaskInfo, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ListAggregatingTasks(queue, group string, opts ...ListOption) ([]*TaskInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
 	opt := composeListOptions(opts...)
 	pgn := rdb.Pagination{Size: opt.pageSize, Page: opt.pageNum - 1}
-	infos, err := i.rdb.ListScheduled(qname, pgn)
+	infos, err := i.rdb.ListAggregating(queue, group, pgn)
+	switch {
+	case errors.IsQueueNotFound(err):
+		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
+	case err != nil:
+		return nil, fmt.Errorf("asynq: %v", err)
+	}
+	var tasks []*TaskInfo
+	for _, i := range infos {
+		tasks = append(tasks, newTaskInfo(
+			i.Message,
+			i.State,
+			i.NextProcessAt,
+			i.Result,
+		))
+	}
+	return tasks, nil
+}
+
+// ListScheduledTasks retrieves scheduled tasks from the specified queue.
+// Tasks are sorted by NextProcessAt in ascending order.
+//
+// By default, it retrieves the first 30 tasks.
+func (i *Inspector) ListScheduledTasks(queue string, opts ...ListOption) ([]*TaskInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
+		return nil, fmt.Errorf("asynq: %v", err)
+	}
+	opt := composeListOptions(opts...)
+	pgn := rdb.Pagination{Size: opt.pageSize, Page: opt.pageNum - 1}
+	infos, err := i.rdb.ListScheduled(queue, pgn)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -365,13 +425,13 @@ func (i *Inspector) ListScheduledTasks(qname string, opts ...ListOption) ([]*Tas
 // Tasks are sorted by NextProcessAt in ascending order.
 //
 // By default, it retrieves the first 30 tasks.
-func (i *Inspector) ListRetryTasks(qname string, opts ...ListOption) ([]*TaskInfo, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ListRetryTasks(queue string, opts ...ListOption) ([]*TaskInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
 	opt := composeListOptions(opts...)
 	pgn := rdb.Pagination{Size: opt.pageSize, Page: opt.pageNum - 1}
-	infos, err := i.rdb.ListRetry(qname, pgn)
+	infos, err := i.rdb.ListRetry(queue, pgn)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -394,13 +454,13 @@ func (i *Inspector) ListRetryTasks(qname string, opts ...ListOption) ([]*TaskInf
 // Tasks are sorted by LastFailedAt in descending order.
 //
 // By default, it retrieves the first 30 tasks.
-func (i *Inspector) ListArchivedTasks(qname string, opts ...ListOption) ([]*TaskInfo, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ListArchivedTasks(queue string, opts ...ListOption) ([]*TaskInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
 	opt := composeListOptions(opts...)
 	pgn := rdb.Pagination{Size: opt.pageSize, Page: opt.pageNum - 1}
-	infos, err := i.rdb.ListArchived(qname, pgn)
+	infos, err := i.rdb.ListArchived(queue, pgn)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -423,13 +483,13 @@ func (i *Inspector) ListArchivedTasks(qname string, opts ...ListOption) ([]*Task
 // Tasks are sorted by expiration time (i.e. CompletedAt + Retention) in descending order.
 //
 // By default, it retrieves the first 30 tasks.
-func (i *Inspector) ListCompletedTasks(qname string, opts ...ListOption) ([]*TaskInfo, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ListCompletedTasks(queue string, opts ...ListOption) ([]*TaskInfo, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return nil, fmt.Errorf("asynq: %v", err)
 	}
 	opt := composeListOptions(opts...)
 	pgn := rdb.Pagination{Size: opt.pageSize, Page: opt.pageNum - 1}
-	infos, err := i.rdb.ListCompleted(qname, pgn)
+	infos, err := i.rdb.ListCompleted(queue, pgn)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return nil, fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -450,51 +510,61 @@ func (i *Inspector) ListCompletedTasks(qname string, opts ...ListOption) ([]*Tas
 
 // DeleteAllPendingTasks deletes all pending tasks from the specified queue,
 // and reports the number tasks deleted.
-func (i *Inspector) DeleteAllPendingTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) DeleteAllPendingTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.DeleteAllPendingTasks(qname)
+	n, err := i.rdb.DeleteAllPendingTasks(queue)
 	return int(n), err
 }
 
 // DeleteAllScheduledTasks deletes all scheduled tasks from the specified queue,
 // and reports the number tasks deleted.
-func (i *Inspector) DeleteAllScheduledTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) DeleteAllScheduledTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.DeleteAllScheduledTasks(qname)
+	n, err := i.rdb.DeleteAllScheduledTasks(queue)
 	return int(n), err
 }
 
 // DeleteAllRetryTasks deletes all retry tasks from the specified queue,
 // and reports the number tasks deleted.
-func (i *Inspector) DeleteAllRetryTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) DeleteAllRetryTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.DeleteAllRetryTasks(qname)
+	n, err := i.rdb.DeleteAllRetryTasks(queue)
 	return int(n), err
 }
 
 // DeleteAllArchivedTasks deletes all archived tasks from the specified queue,
 // and reports the number tasks deleted.
-func (i *Inspector) DeleteAllArchivedTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) DeleteAllArchivedTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.DeleteAllArchivedTasks(qname)
+	n, err := i.rdb.DeleteAllArchivedTasks(queue)
 	return int(n), err
 }
 
 // DeleteAllCompletedTasks deletes all completed tasks from the specified queue,
 // and reports the number tasks deleted.
-func (i *Inspector) DeleteAllCompletedTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) DeleteAllCompletedTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.DeleteAllCompletedTasks(qname)
+	n, err := i.rdb.DeleteAllCompletedTasks(queue)
+	return int(n), err
+}
+
+// DeleteAllAggregatingTasks deletes all tasks from the specified group,
+// and reports the number of tasks deleted.
+func (i *Inspector) DeleteAllAggregatingTasks(queue, group string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
+		return 0, err
+	}
+	n, err := i.rdb.DeleteAllAggregatingTasks(queue, group)
 	return int(n), err
 }
 
@@ -505,11 +575,11 @@ func (i *Inspector) DeleteAllCompletedTasks(qname string) (int, error) {
 // If a queue with the given name doesn't exist, it returns an error wrapping ErrQueueNotFound.
 // If a task with the given id doesn't exist in the queue, it returns an error wrapping ErrTaskNotFound.
 // If the task is in active state, it returns a non-nil error.
-func (i *Inspector) DeleteTask(qname, id string) error {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) DeleteTask(queue, id string) error {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return fmt.Errorf("asynq: %v", err)
 	}
-	err := i.rdb.DeleteTask(qname, id)
+	err := i.rdb.DeleteTask(queue, id)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -522,33 +592,43 @@ func (i *Inspector) DeleteTask(qname, id string) error {
 
 }
 
-// RunAllScheduledTasks transition all scheduled tasks to pending state from the given queue,
-// and reports the number of tasks transitioned.
-func (i *Inspector) RunAllScheduledTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+// RunAllScheduledTasks schedules all scheduled tasks from the given queue to run,
+// and reports the number of tasks scheduled to run.
+func (i *Inspector) RunAllScheduledTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.RunAllScheduledTasks(qname)
+	n, err := i.rdb.RunAllScheduledTasks(queue)
 	return int(n), err
 }
 
-// RunAllRetryTasks transition all retry tasks to pending state from the given queue,
-// and reports the number of tasks transitioned.
-func (i *Inspector) RunAllRetryTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+// RunAllRetryTasks schedules all retry tasks from the given queue to run,
+// and reports the number of tasks scheduled to run.
+func (i *Inspector) RunAllRetryTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.RunAllRetryTasks(qname)
+	n, err := i.rdb.RunAllRetryTasks(queue)
 	return int(n), err
 }
 
-// RunAllArchivedTasks transition all archived tasks to pending state from the given queue,
-// and reports the number of tasks transitioned.
-func (i *Inspector) RunAllArchivedTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+// RunAllArchivedTasks schedules all archived tasks from the given queue to run,
+// and reports the number of tasks scheduled to run.
+func (i *Inspector) RunAllArchivedTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.RunAllArchivedTasks(qname)
+	n, err := i.rdb.RunAllArchivedTasks(queue)
+	return int(n), err
+}
+
+// RunAllAggregatingTasks schedules all tasks from the given grou to run.
+// and reports the number of tasks scheduled to run.
+func (i *Inspector) RunAllAggregatingTasks(queue, group string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
+		return 0, err
+	}
+	n, err := i.rdb.RunAllAggregatingTasks(queue, group)
 	return int(n), err
 }
 
@@ -559,11 +639,11 @@ func (i *Inspector) RunAllArchivedTasks(qname string) (int, error) {
 // If a queue with the given name doesn't exist, it returns an error wrapping ErrQueueNotFound.
 // If a task with the given id doesn't exist in the queue, it returns an error wrapping ErrTaskNotFound.
 // If the task is in pending or active state, it returns a non-nil error.
-func (i *Inspector) RunTask(qname, id string) error {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) RunTask(queue, id string) error {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return fmt.Errorf("asynq: %v", err)
 	}
-	err := i.rdb.RunTask(qname, id)
+	err := i.rdb.RunTask(queue, id)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -577,31 +657,41 @@ func (i *Inspector) RunTask(qname, id string) error {
 
 // ArchiveAllPendingTasks archives all pending tasks from the given queue,
 // and reports the number of tasks archived.
-func (i *Inspector) ArchiveAllPendingTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ArchiveAllPendingTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.ArchiveAllPendingTasks(qname)
+	n, err := i.rdb.ArchiveAllPendingTasks(queue)
 	return int(n), err
 }
 
 // ArchiveAllScheduledTasks archives all scheduled tasks from the given queue,
 // and reports the number of tasks archiveed.
-func (i *Inspector) ArchiveAllScheduledTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ArchiveAllScheduledTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.ArchiveAllScheduledTasks(qname)
+	n, err := i.rdb.ArchiveAllScheduledTasks(queue)
 	return int(n), err
 }
 
 // ArchiveAllRetryTasks archives all retry tasks from the given queue,
 // and reports the number of tasks archiveed.
-func (i *Inspector) ArchiveAllRetryTasks(qname string) (int, error) {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ArchiveAllRetryTasks(queue string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return 0, err
 	}
-	n, err := i.rdb.ArchiveAllRetryTasks(qname)
+	n, err := i.rdb.ArchiveAllRetryTasks(queue)
+	return int(n), err
+}
+
+// ArchiveAllAggregatingTasks archives all tasks from the given group,
+// and reports the number of tasks archived.
+func (i *Inspector) ArchiveAllAggregatingTasks(queue, group string) (int, error) {
+	if err := base.ValidateQueueName(queue); err != nil {
+		return 0, err
+	}
+	n, err := i.rdb.ArchiveAllAggregatingTasks(queue, group)
 	return int(n), err
 }
 
@@ -612,11 +702,11 @@ func (i *Inspector) ArchiveAllRetryTasks(qname string) (int, error) {
 // If a queue with the given name doesn't exist, it returns an error wrapping ErrQueueNotFound.
 // If a task with the given id doesn't exist in the queue, it returns an error wrapping ErrTaskNotFound.
 // If the task is in already archived, it returns a non-nil error.
-func (i *Inspector) ArchiveTask(qname, id string) error {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) ArchiveTask(queue, id string) error {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return fmt.Errorf("asynq: err")
 	}
-	err := i.rdb.ArchiveTask(qname, id)
+	err := i.rdb.ArchiveTask(queue, id)
 	switch {
 	case errors.IsQueueNotFound(err):
 		return fmt.Errorf("asynq: %w", ErrQueueNotFound)
@@ -638,20 +728,20 @@ func (i *Inspector) CancelProcessing(id string) error {
 
 // PauseQueue pauses task processing on the specified queue.
 // If the queue is already paused, it will return a non-nil error.
-func (i *Inspector) PauseQueue(qname string) error {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) PauseQueue(queue string) error {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return err
 	}
-	return i.rdb.Pause(qname)
+	return i.rdb.Pause(queue)
 }
 
 // UnpauseQueue resumes task processing on the specified queue.
 // If the queue is not paused, it will return a non-nil error.
-func (i *Inspector) UnpauseQueue(qname string) error {
-	if err := base.ValidateQueueName(qname); err != nil {
+func (i *Inspector) UnpauseQueue(queue string) error {
+	if err := base.ValidateQueueName(queue); err != nil {
 		return err
 	}
-	return i.rdb.Unpause(qname)
+	return i.rdb.Unpause(queue)
 }
 
 // Servers return a list of running servers' information.
@@ -741,8 +831,8 @@ type WorkerInfo struct {
 }
 
 // ClusterKeySlot returns an integer identifying the hash slot the given queue hashes to.
-func (i *Inspector) ClusterKeySlot(qname string) (int64, error) {
-	return i.rdb.ClusterKeySlot(qname)
+func (i *Inspector) ClusterKeySlot(queue string) (int64, error) {
+	return i.rdb.ClusterKeySlot(queue)
 }
 
 // ClusterNode describes a node in redis cluster.
@@ -757,8 +847,8 @@ type ClusterNode struct {
 // ClusterNodes returns a list of nodes the given queue belongs to.
 //
 // Only relevant if task queues are stored in redis cluster.
-func (i *Inspector) ClusterNodes(qname string) ([]*ClusterNode, error) {
-	nodes, err := i.rdb.ClusterNodes(qname)
+func (i *Inspector) ClusterNodes(queue string) ([]*ClusterNode, error) {
+	nodes, err := i.rdb.ClusterNodes(queue)
 	if err != nil {
 		return nil, err
 	}
@@ -826,11 +916,11 @@ func parseOption(s string) (Option, error) {
 	fn, arg := parseOptionFunc(s), parseOptionArg(s)
 	switch fn {
 	case "Queue":
-		qname, err := strconv.Unquote(arg)
+		queue, err := strconv.Unquote(arg)
 		if err != nil {
 			return nil, err
 		}
-		return Queue(qname), nil
+		return Queue(queue), nil
 	case "MaxRetry":
 		n, err := strconv.Atoi(arg)
 		if err != nil {

inspector_test.go

@@ -12,12 +12,13 @@ import (
 	"testing"
 	"time"
 
+	"github.com/go-redis/redis/v8"
 	"github.com/google/go-cmp/cmp"
 	"github.com/google/go-cmp/cmp/cmpopts"
 	"github.com/google/uuid"
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/rdb"
+	h "github.com/hibiken/asynq/internal/testutil"
 	"github.com/hibiken/asynq/internal/timeutil"
 )
 
@@ -1120,6 +1121,106 @@ func TestInspectorListCompletedTasks(t *testing.T) {
 	}
 }
 
+func TestInspectorListAggregatingTasks(t *testing.T) {
+	r := setup(t)
+	defer r.Close()
+	now := time.Now()
+
+	m1 := h.NewTaskMessageBuilder().SetType("task1").SetQueue("default").SetGroup("group1").Build()
+	m2 := h.NewTaskMessageBuilder().SetType("task2").SetQueue("default").SetGroup("group1").Build()
+	m3 := h.NewTaskMessageBuilder().SetType("task3").SetQueue("default").SetGroup("group1").Build()
+	m4 := h.NewTaskMessageBuilder().SetType("task4").SetQueue("default").SetGroup("group2").Build()
+	m5 := h.NewTaskMessageBuilder().SetType("task5").SetQueue("custom").SetGroup("group1").Build()
+
+	inspector := NewInspector(getRedisConnOpt(t))
+
+	fxt := struct {
+		tasks     []*h.TaskSeedData
+		allQueues []string
+		allGroups map[string][]string
+		groups    map[string][]*redis.Z
+	}{
+		tasks: []*h.TaskSeedData{
+			{Msg: m1, State: base.TaskStateAggregating},
+			{Msg: m2, State: base.TaskStateAggregating},
+			{Msg: m3, State: base.TaskStateAggregating},
+			{Msg: m4, State: base.TaskStateAggregating},
+			{Msg: m5, State: base.TaskStateAggregating},
+		},
+		allQueues: []string{"default", "custom"},
+		allGroups: map[string][]string{
+			base.AllGroups("default"): {"group1", "group2"},
+			base.AllGroups("custom"):  {"group1"},
+		},
+		groups: map[string][]*redis.Z{
+			base.GroupKey("default", "group1"): {
+				{Member: m1.ID, Score: float64(now.Add(-30 * time.Second).Unix())},
+				{Member: m2.ID, Score: float64(now.Add(-20 * time.Second).Unix())},
+				{Member: m3.ID, Score: float64(now.Add(-10 * time.Second).Unix())},
+			},
+			base.GroupKey("default", "group2"): {
+				{Member: m4.ID, Score: float64(now.Add(-30 * time.Second).Unix())},
+			},
+			base.GroupKey("custom", "group1"): {
+				{Member: m5.ID, Score: float64(now.Add(-30 * time.Second).Unix())},
+			},
+		},
+	}
+
+	tests := []struct {
+		desc  string
+		qname string
+		gname string
+		want  []*TaskInfo
+	}{
+		{
+			desc:  "default queue group1",
+			qname: "default",
+			gname: "group1",
+			want: []*TaskInfo{
+				createAggregatingTaskInfo(m1),
+				createAggregatingTaskInfo(m2),
+				createAggregatingTaskInfo(m3),
+			},
+		},
+		{
+			desc:  "custom queue group1",
+			qname: "custom",
+			gname: "group1",
+			want: []*TaskInfo{
+				createAggregatingTaskInfo(m5),
+			},
+		},
+	}
+
+	for _, tc := range tests {
+		h.FlushDB(t, r)
+		h.SeedTasks(t, r, fxt.tasks)
+		h.SeedRedisSet(t, r, base.AllQueues, fxt.allQueues)
+		h.SeedRedisSets(t, r, fxt.allGroups)
+		h.SeedRedisZSets(t, r, fxt.groups)
+
+		t.Run(tc.desc, func(t *testing.T) {
+			got, err := inspector.ListAggregatingTasks(tc.qname, tc.gname)
+			if err != nil {
+				t.Fatalf("ListAggregatingTasks returned error: %v", err)
+			}
+
+			cmpOpts := []cmp.Option{
+				cmpopts.EquateApproxTime(2 * time.Second),
+				cmp.AllowUnexported(TaskInfo{}),
+			}
+			if diff := cmp.Diff(tc.want, got, cmpOpts...); diff != "" {
+				t.Errorf("ListAggregatingTasks = %v, want = %v; (-want,+got)\n%s", got, tc.want, diff)
+			}
+		})
+	}
+}
+
+func createAggregatingTaskInfo(msg *base.TaskMessage) *TaskInfo {
+	return newTaskInfo(msg, base.TaskStateAggregating, time.Time{}, nil)
+}
+
 func TestInspectorListPagination(t *testing.T) {
 	// Create 100 tasks.
 	var msgs []*base.TaskMessage
@@ -1221,6 +1322,9 @@ func TestInspectorListTasksQueueNotFoundError(t *testing.T) {
 		if _, err := inspector.ListCompletedTasks(tc.qname); !errors.Is(err, tc.wantErr) {
 			t.Errorf("ListCompletedTasks(%q) returned error %v, want %v", tc.qname, err, tc.wantErr)
 		}
+		if _, err := inspector.ListAggregatingTasks(tc.qname, "mygroup"); !errors.Is(err, tc.wantErr) {
+			t.Errorf("ListAggregatingTasks(%q, \"mygroup\") returned error %v, want %v", tc.qname, err, tc.wantErr)
+		}
 	}
 }
 
@@ -3323,3 +3427,99 @@ func TestParseOption(t *testing.T) {
 		})
 	}
 }
+
+func TestInspectorGroups(t *testing.T) {
+	r := setup(t)
+	defer r.Close()
+	inspector := NewInspector(getRedisConnOpt(t))
+
+	m1 := h.NewTaskMessageBuilder().SetGroup("group1").Build()
+	m2 := h.NewTaskMessageBuilder().SetGroup("group1").Build()
+	m3 := h.NewTaskMessageBuilder().SetGroup("group1").Build()
+	m4 := h.NewTaskMessageBuilder().SetGroup("group2").Build()
+	m5 := h.NewTaskMessageBuilder().SetQueue("custom").SetGroup("group1").Build()
+	m6 := h.NewTaskMessageBuilder().SetQueue("custom").SetGroup("group1").Build()
+
+	now := time.Now()
+
+	fixtures := struct {
+		tasks     []*h.TaskSeedData
+		allGroups map[string][]string
+		groups    map[string][]*redis.Z
+	}{
+		tasks: []*h.TaskSeedData{
+			{Msg: m1, State: base.TaskStateAggregating},
+			{Msg: m2, State: base.TaskStateAggregating},
+			{Msg: m3, State: base.TaskStateAggregating},
+			{Msg: m4, State: base.TaskStateAggregating},
+			{Msg: m5, State: base.TaskStateAggregating},
+		},
+		allGroups: map[string][]string{
+			base.AllGroups("default"): {"group1", "group2"},
+			base.AllGroups("custom"):  {"group1"},
+		},
+		groups: map[string][]*redis.Z{
+			base.GroupKey("default", "group1"): {
+				{Member: m1.ID, Score: float64(now.Add(-10 * time.Second).Unix())},
+				{Member: m2.ID, Score: float64(now.Add(-20 * time.Second).Unix())},
+				{Member: m3.ID, Score: float64(now.Add(-30 * time.Second).Unix())},
+			},
+			base.GroupKey("default", "group2"): {
+				{Member: m4.ID, Score: float64(now.Add(-20 * time.Second).Unix())},
+			},
+			base.GroupKey("custom", "group1"): {
+				{Member: m5.ID, Score: float64(now.Add(-10 * time.Second).Unix())},
+				{Member: m6.ID, Score: float64(now.Add(-20 * time.Second).Unix())},
+			},
+		},
+	}
+
+	tests := []struct {
+		desc  string
+		qname string
+		want  []*GroupInfo
+	}{
+		{
+			desc:  "default queue groups",
+			qname: "default",
+			want: []*GroupInfo{
+				{Group: "group1", Size: 3},
+				{Group: "group2", Size: 1},
+			},
+		},
+		{
+			desc:  "custom queue groups",
+			qname: "custom",
+			want: []*GroupInfo{
+				{Group: "group1", Size: 2},
+			},
+		},
+	}
+
+	var sortGroupInfosOpt = cmp.Transformer(
+		"SortGroupInfos",
+		func(in []*GroupInfo) []*GroupInfo {
+			out := append([]*GroupInfo(nil), in...)
+			sort.Slice(out, func(i, j int) bool {
+				return out[i].Group < out[j].Group
+			})
+			return out
+		})
+
+	for _, tc := range tests {
+		h.FlushDB(t, r)
+		h.SeedTasks(t, r, fixtures.tasks)
+		h.SeedRedisSets(t, r, fixtures.allGroups)
+		h.SeedRedisZSets(t, r, fixtures.groups)
+
+		t.Run(tc.desc, func(t *testing.T) {
+			got, err := inspector.Groups(tc.qname)
+			if err != nil {
+				t.Fatalf("Groups returned error: %v", err)
+			}
+			if diff := cmp.Diff(tc.want, got, sortGroupInfosOpt); diff != "" {
+				t.Errorf("Groups = %v, want %v; (-want,+got)\n%s", got, tc.want, diff)
+			}
+		})
+	}
+}

internal/base/base.go

@@ -23,7 +23,7 @@ import (
 )
 
 // Version of asynq library and CLI.
-const Version = "0.22.1"
+const Version = "0.23.0"
 
 // DefaultQueueName is the queue name used if none are specified by user.
 const DefaultQueueName = "default"
@@ -50,6 +50,7 @@ const (
 	TaskStateRetry
 	TaskStateArchived
 	TaskStateCompleted
+	TaskStateAggregating // describes a state where task is waiting in a group to be aggregated
 )
 
 func (s TaskState) String() string {
@@ -66,6 +67,8 @@ func (s TaskState) String() string {
 		return "archived"
 	case TaskStateCompleted:
 		return "completed"
+	case TaskStateAggregating:
+		return "aggregating"
 	}
 	panic(fmt.Sprintf("internal error: unknown task state %d", s))
 }
@@ -84,6 +87,8 @@ func TaskStateFromString(s string) (TaskState, error) {
 		return TaskStateArchived, nil
 	case "completed":
 		return TaskStateCompleted, nil
+	case "aggregating":
+		return TaskStateAggregating, nil
 	}
 	return 0, errors.E(errors.FailedPrecondition, fmt.Sprintf("%q is not supported task state", s))
 }
@@ -200,6 +205,32 @@ func UniqueKey(qname, tasktype string, payload []byte) string {
 	return fmt.Sprintf("%sunique:%s:%s", QueueKeyPrefix(qname), tasktype, hex.EncodeToString(checksum[:]))
 }
 
+// GroupKeyPrefix returns a prefix for group key.
+func GroupKeyPrefix(qname string) string {
+	return fmt.Sprintf("%sg:", QueueKeyPrefix(qname))
+}
+
+// GroupKey returns a redis key used to group tasks belong in the same group.
+func GroupKey(qname, gkey string) string {
+	return fmt.Sprintf("%s%s", GroupKeyPrefix(qname), gkey)
+}
+
+// AggregationSetKey returns a redis key used for an aggregation set.
+func AggregationSetKey(qname, gname, setID string) string {
+	return fmt.Sprintf("%s:%s", GroupKey(qname, gname), setID)
+}
+
+// AllGroups return a redis key used to store all group keys used in a given queue.
+func AllGroups(qname string) string {
+	return fmt.Sprintf("%sgroups", QueueKeyPrefix(qname))
+}
+
+// AllAggregationSets returns a redis key used to store all aggregation sets (set of tasks staged to be aggregated)
+// in a given queue.
+func AllAggregationSets(qname string) string {
+	return fmt.Sprintf("%saggregation_sets", QueueKeyPrefix(qname))
+}
+
 // TaskMessage is the internal representation of a task with additional metadata fields.
 // Serialized data of this type gets written to redis.
 type TaskMessage struct {
@@ -250,6 +281,11 @@ type TaskMessage struct {
 	// Empty string indicates that no uniqueness lock was used.
 	UniqueKey string
 
+	// GroupKey holds the group key used for task aggregation.
+	//
+	// Empty string indicates no aggregation is used for this task.
+	GroupKey string
+
 	// Retention specifies the number of seconds the task should be retained after completion.
 	Retention int64
 
@@ -277,6 +313,7 @@ func EncodeMessage(msg *TaskMessage) ([]byte, error) {
 		Timeout:      msg.Timeout,
 		Deadline:     msg.Deadline,
 		UniqueKey:    msg.UniqueKey,
+		GroupKey:     msg.GroupKey,
 		Retention:    msg.Retention,
 		CompletedAt:  msg.CompletedAt,
 	})
@@ -300,6 +337,7 @@ func DecodeMessage(data []byte) (*TaskMessage, error) {
 		Timeout:      pbmsg.GetTimeout(),
 		Deadline:     pbmsg.GetDeadline(),
 		UniqueKey:    pbmsg.GetUniqueKey(),
+		GroupKey:     pbmsg.GetGroupKey(),
 		Retention:    pbmsg.GetRetention(),
 		CompletedAt:  pbmsg.GetCompletedAt(),
 	}, nil
@@ -676,6 +714,7 @@ func (l *Lease) IsValid() bool {
 // See rdb.RDB as a reference implementation.
 type Broker interface {
 	Ping() error
+	Close() error
 	Enqueue(ctx context.Context, msg *TaskMessage) error
 	EnqueueUnique(ctx context.Context, msg *TaskMessage, ttl time.Duration) error
 	Dequeue(qnames ...string) (*TaskMessage, time.Time, error)
@@ -687,13 +726,30 @@ type Broker interface {
 	Retry(ctx context.Context, msg *TaskMessage, processAt time.Time, errMsg string, isFailure bool) error
 	Archive(ctx context.Context, msg *TaskMessage, errMsg string) error
 	ForwardIfReady(qnames ...string) error
+
+	// Group aggregation related methods
+	AddToGroup(ctx context.Context, msg *TaskMessage, gname string) error
+	AddToGroupUnique(ctx context.Context, msg *TaskMessage, groupKey string, ttl time.Duration) error
+	ListGroups(qname string) ([]string, error)
+	AggregationCheck(qname, gname string, t time.Time, gracePeriod, maxDelay time.Duration, maxSize int) (aggregationSetID string, err error)
+	ReadAggregationSet(qname, gname, aggregationSetID string) ([]*TaskMessage, time.Time, error)
+	DeleteAggregationSet(ctx context.Context, qname, gname, aggregationSetID string) error
+	ReclaimStaleAggregationSets(qname string) error
+
+	// Task retention related method
 	DeleteExpiredCompletedTasks(qname string) error
+
+	// Lease related methods
 	ListLeaseExpired(cutoff time.Time, qnames ...string) ([]*TaskMessage, error)
 	ExtendLease(qname string, ids ...string) (time.Time, error)
+
+	// State snapshot related methods
 	WriteServerState(info *ServerInfo, workers []*WorkerInfo, ttl time.Duration) error
 	ClearServerState(host string, pid int, serverID string) error
+
+	// Cancelation related methods
 	CancelationPubSub() (*redis.PubSub, error) // TODO: Need to decouple from redis to support other brokers
 	PublishCancelation(id string) error
+
 	WriteResult(qname, id string, data []byte) (n int, err error)
-	Close() error
 }

internal/base/base_test.go

@@ -395,6 +395,107 @@ func TestUniqueKey(t *testing.T) {
 	}
 }
 
+func TestGroupKey(t *testing.T) {
+	tests := []struct {
+		qname string
+		gkey  string
+		want  string
+	}{
+		{
+			qname: "default",
+			gkey:  "mygroup",
+			want:  "asynq:{default}:g:mygroup",
+		},
+		{
+			qname: "custom",
+			gkey:  "foo",
+			want:  "asynq:{custom}:g:foo",
+		},
+	}
+
+	for _, tc := range tests {
+		got := GroupKey(tc.qname, tc.gkey)
+		if got != tc.want {
+			t.Errorf("GroupKey(%q, %q) = %q, want %q", tc.qname, tc.gkey, got, tc.want)
+		}
+	}
+}
+
+func TestAggregationSetKey(t *testing.T) {
+	tests := []struct {
+		qname string
+		gname string
+		setID string
+		want  string
+	}{
+		{
+			qname: "default",
+			gname: "mygroup",
+			setID: "12345",
+			want:  "asynq:{default}:g:mygroup:12345",
+		},
+		{
+			qname: "custom",
+			gname: "foo",
+			setID: "98765",
+			want:  "asynq:{custom}:g:foo:98765",
+		},
+	}
+
+	for _, tc := range tests {
+		got := AggregationSetKey(tc.qname, tc.gname, tc.setID)
+		if got != tc.want {
+			t.Errorf("AggregationSetKey(%q, %q, %q) = %q, want %q", tc.qname, tc.gname, tc.setID, got, tc.want)
+		}
+	}
+}
+
+func TestAllGroups(t *testing.T) {
+	tests := []struct {
+		qname string
+		want  string
+	}{
+		{
+			qname: "default",
+			want:  "asynq:{default}:groups",
+		},
+		{
+			qname: "custom",
+			want:  "asynq:{custom}:groups",
+		},
+	}
+
+	for _, tc := range tests {
+		got := AllGroups(tc.qname)
+		if got != tc.want {
+			t.Errorf("AllGroups(%q) = %q, want %q", tc.qname, got, tc.want)
+		}
+	}
+}
+
+func TestAllAggregationSets(t *testing.T) {
+	tests := []struct {
+		qname string
+		want  string
+	}{
+		{
+			qname: "default",
+			want:  "asynq:{default}:aggregation_sets",
+		},
+		{
+			qname: "custom",
+			want:  "asynq:{custom}:aggregation_sets",
+		},
+	}
+
+	for _, tc := range tests {
+		got := AllAggregationSets(tc.qname)
+		if got != tc.want {
+			t.Errorf("AllAggregationSets(%q) = %q, want %q", tc.qname, got, tc.want)
+		}
+	}
+}
+
 func TestMessageEncoding(t *testing.T) {
 	id := uuid.NewString()
 	tests := []struct {
@@ -407,6 +508,7 @@ func TestMessageEncoding(t *testing.T) {
 				Payload:   toBytes(map[string]interface{}{"a": 1, "b": "hello!", "c": true}),
 				ID:        id,
 				Queue:     "default",
+				GroupKey:  "mygroup",
 				Retry:     10,
 				Retried:   0,
 				Timeout:   1800,
@@ -418,6 +520,7 @@ func TestMessageEncoding(t *testing.T) {
 				Payload:   toBytes(map[string]interface{}{"a": json.Number("1"), "b": "hello!", "c": true}),
 				ID:        id,
 				Queue:     "default",
+				GroupKey:  "mygroup",
 				Retry:     10,
 				Retried:   0,
 				Timeout:   1800,

internal/proto/asynq.pb.go

@@ -32,6 +32,7 @@ const _ = proto.ProtoPackageIsVersion4
 
 // TaskMessage is the internal representation of a task with additional
 // metadata fields.
+// Next ID: 15
 type TaskMessage struct {
 	state         protoimpl.MessageState
 	sizeCache     protoimpl.SizeCache
@@ -65,6 +66,9 @@ type TaskMessage struct {
 	// UniqueKey holds the redis key used for uniqueness lock for this task.
 	// Empty string indicates that no uniqueness lock was used.
 	UniqueKey string `protobuf:"bytes,10,opt,name=unique_key,json=uniqueKey,proto3" json:"unique_key,omitempty"`
+	// GroupKey is a name of the group used for task aggregation.
+	// This field is optional and empty value means no aggregation for the task.
+	GroupKey string `protobuf:"bytes,14,opt,name=group_key,json=groupKey,proto3" json:"group_key,omitempty"`
 	// Retention period specified in a number of seconds.
 	// The task will be stored in redis as a completed task until the TTL
 	// expires.
@@ -184,6 +188,13 @@ func (x *TaskMessage) GetUniqueKey() string {
 	return ""
 }
 
+func (x *TaskMessage) GetGroupKey() string {
+	if x != nil {
+		return x.GroupKey
+	}
+	return ""
+}
+
 func (x *TaskMessage) GetRetention() int64 {
 	if x != nil {
 		return x.Retention
@@ -614,7 +625,7 @@ var file_asynq_proto_rawDesc = []byte{
 	0x0a, 0x0b, 0x61, 0x73, 0x79, 0x6e, 0x71, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x12, 0x05, 0x61,
 	0x73, 0x79, 0x6e, 0x71, 0x1a, 0x1f, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2f, 0x70, 0x72, 0x6f,
 	0x74, 0x6f, 0x62, 0x75, 0x66, 0x2f, 0x74, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x2e,
-	0x70, 0x72, 0x6f, 0x74, 0x6f, 0x22, 0xea, 0x02, 0x0a, 0x0b, 0x54, 0x61, 0x73, 0x6b, 0x4d, 0x65,
+	0x70, 0x72, 0x6f, 0x74, 0x6f, 0x22, 0x87, 0x03, 0x0a, 0x0b, 0x54, 0x61, 0x73, 0x6b, 0x4d, 0x65,
 	0x73, 0x73, 0x61, 0x67, 0x65, 0x12, 0x12, 0x0a, 0x04, 0x74, 0x79, 0x70, 0x65, 0x18, 0x01, 0x20,
 	0x01, 0x28, 0x09, 0x52, 0x04, 0x74, 0x79, 0x70, 0x65, 0x12, 0x18, 0x0a, 0x07, 0x70, 0x61, 0x79,
 	0x6c, 0x6f, 0x61, 0x64, 0x18, 0x02, 0x20, 0x01, 0x28, 0x0c, 0x52, 0x07, 0x70, 0x61, 0x79, 0x6c,
@@ -633,84 +644,86 @@ var file_asynq_proto_rawDesc = []byte{
 	0x6e, 0x65, 0x18, 0x09, 0x20, 0x01, 0x28, 0x03, 0x52, 0x08, 0x64, 0x65, 0x61, 0x64, 0x6c, 0x69,
 	0x6e, 0x65, 0x12, 0x1d, 0x0a, 0x0a, 0x75, 0x6e, 0x69, 0x71, 0x75, 0x65, 0x5f, 0x6b, 0x65, 0x79,
 	0x18, 0x0a, 0x20, 0x01, 0x28, 0x09, 0x52, 0x09, 0x75, 0x6e, 0x69, 0x71, 0x75, 0x65, 0x4b, 0x65,
-	0x79, 0x12, 0x1c, 0x0a, 0x09, 0x72, 0x65, 0x74, 0x65, 0x6e, 0x74, 0x69, 0x6f, 0x6e, 0x18, 0x0c,
-	0x20, 0x01, 0x28, 0x03, 0x52, 0x09, 0x72, 0x65, 0x74, 0x65, 0x6e, 0x74, 0x69, 0x6f, 0x6e, 0x12,
-	0x21, 0x0a, 0x0c, 0x63, 0x6f, 0x6d, 0x70, 0x6c, 0x65, 0x74, 0x65, 0x64, 0x5f, 0x61, 0x74, 0x18,
-	0x0d, 0x20, 0x01, 0x28, 0x03, 0x52, 0x0b, 0x63, 0x6f, 0x6d, 0x70, 0x6c, 0x65, 0x74, 0x65, 0x64,
-	0x41, 0x74, 0x22, 0x8f, 0x03, 0x0a, 0x0a, 0x53, 0x65, 0x72, 0x76, 0x65, 0x72, 0x49, 0x6e, 0x66,
-	0x6f, 0x12, 0x12, 0x0a, 0x04, 0x68, 0x6f, 0x73, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52,
-	0x04, 0x68, 0x6f, 0x73, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x70, 0x69, 0x64, 0x18, 0x02, 0x20, 0x01,
-	0x28, 0x05, 0x52, 0x03, 0x70, 0x69, 0x64, 0x12, 0x1b, 0x0a, 0x09, 0x73, 0x65, 0x72, 0x76, 0x65,
-	0x72, 0x5f, 0x69, 0x64, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x73, 0x65, 0x72, 0x76,
-	0x65, 0x72, 0x49, 0x64, 0x12, 0x20, 0x0a, 0x0b, 0x63, 0x6f, 0x6e, 0x63, 0x75, 0x72, 0x72, 0x65,
-	0x6e, 0x63, 0x79, 0x18, 0x04, 0x20, 0x01, 0x28, 0x05, 0x52, 0x0b, 0x63, 0x6f, 0x6e, 0x63, 0x75,
-	0x72, 0x72, 0x65, 0x6e, 0x63, 0x79, 0x12, 0x35, 0x0a, 0x06, 0x71, 0x75, 0x65, 0x75, 0x65, 0x73,
-	0x18, 0x05, 0x20, 0x03, 0x28, 0x0b, 0x32, 0x1d, 0x2e, 0x61, 0x73, 0x79, 0x6e, 0x71, 0x2e, 0x53,
-	0x65, 0x72, 0x76, 0x65, 0x72, 0x49, 0x6e, 0x66, 0x6f, 0x2e, 0x51, 0x75, 0x65, 0x75, 0x65, 0x73,
-	0x45, 0x6e, 0x74, 0x72, 0x79, 0x52, 0x06, 0x71, 0x75, 0x65, 0x75, 0x65, 0x73, 0x12, 0x27, 0x0a,
-	0x0f, 0x73, 0x74, 0x72, 0x69, 0x63, 0x74, 0x5f, 0x70, 0x72, 0x69, 0x6f, 0x72, 0x69, 0x74, 0x79,
-	0x18, 0x06, 0x20, 0x01, 0x28, 0x08, 0x52, 0x0e, 0x73, 0x74, 0x72, 0x69, 0x63, 0x74, 0x50, 0x72,
-	0x69, 0x6f, 0x72, 0x69, 0x74, 0x79, 0x12, 0x16, 0x0a, 0x06, 0x73, 0x74, 0x61, 0x74, 0x75, 0x73,
-	0x18, 0x07, 0x20, 0x01, 0x28, 0x09, 0x52, 0x06, 0x73, 0x74, 0x61, 0x74, 0x75, 0x73, 0x12, 0x39,
-	0x0a, 0x0a, 0x73, 0x74, 0x61, 0x72, 0x74, 0x5f, 0x74, 0x69, 0x6d, 0x65, 0x18, 0x08, 0x20, 0x01,
-	0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74,
-	0x6f, 0x62, 0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x09,
-	0x73, 0x74, 0x61, 0x72, 0x74, 0x54, 0x69, 0x6d, 0x65, 0x12, 0x2e, 0x0a, 0x13, 0x61, 0x63, 0x74,
-	0x69, 0x76, 0x65, 0x5f, 0x77, 0x6f, 0x72, 0x6b, 0x65, 0x72, 0x5f, 0x63, 0x6f, 0x75, 0x6e, 0x74,
-	0x18, 0x09, 0x20, 0x01, 0x28, 0x05, 0x52, 0x11, 0x61, 0x63, 0x74, 0x69, 0x76, 0x65, 0x57, 0x6f,
-	0x72, 0x6b, 0x65, 0x72, 0x43, 0x6f, 0x75, 0x6e, 0x74, 0x1a, 0x39, 0x0a, 0x0b, 0x51, 0x75, 0x65,
-	0x75, 0x65, 0x73, 0x45, 0x6e, 0x74, 0x72, 0x79, 0x12, 0x10, 0x0a, 0x03, 0x6b, 0x65, 0x79, 0x18,
-	0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x03, 0x6b, 0x65, 0x79, 0x12, 0x14, 0x0a, 0x05, 0x76, 0x61,
-	0x6c, 0x75, 0x65, 0x18, 0x02, 0x20, 0x01, 0x28, 0x05, 0x52, 0x05, 0x76, 0x61, 0x6c, 0x75, 0x65,
-	0x3a, 0x02, 0x38, 0x01, 0x22, 0xb1, 0x02, 0x0a, 0x0a, 0x57, 0x6f, 0x72, 0x6b, 0x65, 0x72, 0x49,
-	0x6e, 0x66, 0x6f, 0x12, 0x12, 0x0a, 0x04, 0x68, 0x6f, 0x73, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28,
-	0x09, 0x52, 0x04, 0x68, 0x6f, 0x73, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x70, 0x69, 0x64, 0x18, 0x02,
-	0x20, 0x01, 0x28, 0x05, 0x52, 0x03, 0x70, 0x69, 0x64, 0x12, 0x1b, 0x0a, 0x09, 0x73, 0x65, 0x72,
-	0x76, 0x65, 0x72, 0x5f, 0x69, 0x64, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x73, 0x65,
-	0x72, 0x76, 0x65, 0x72, 0x49, 0x64, 0x12, 0x17, 0x0a, 0x07, 0x74, 0x61, 0x73, 0x6b, 0x5f, 0x69,
-	0x64, 0x18, 0x04, 0x20, 0x01, 0x28, 0x09, 0x52, 0x06, 0x74, 0x61, 0x73, 0x6b, 0x49, 0x64, 0x12,
-	0x1b, 0x0a, 0x09, 0x74, 0x61, 0x73, 0x6b, 0x5f, 0x74, 0x79, 0x70, 0x65, 0x18, 0x05, 0x20, 0x01,
-	0x28, 0x09, 0x52, 0x08, 0x74, 0x61, 0x73, 0x6b, 0x54, 0x79, 0x70, 0x65, 0x12, 0x21, 0x0a, 0x0c,
-	0x74, 0x61, 0x73, 0x6b, 0x5f, 0x70, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x18, 0x06, 0x20, 0x01,
-	0x28, 0x0c, 0x52, 0x0b, 0x74, 0x61, 0x73, 0x6b, 0x50, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x12,
-	0x14, 0x0a, 0x05, 0x71, 0x75, 0x65, 0x75, 0x65, 0x18, 0x07, 0x20, 0x01, 0x28, 0x09, 0x52, 0x05,
-	0x71, 0x75, 0x65, 0x75, 0x65, 0x12, 0x39, 0x0a, 0x0a, 0x73, 0x74, 0x61, 0x72, 0x74, 0x5f, 0x74,
-	0x69, 0x6d, 0x65, 0x18, 0x08, 0x20, 0x01, 0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67,
-	0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65,
-	0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x09, 0x73, 0x74, 0x61, 0x72, 0x74, 0x54, 0x69, 0x6d, 0x65,
-	0x12, 0x36, 0x0a, 0x08, 0x64, 0x65, 0x61, 0x64, 0x6c, 0x69, 0x6e, 0x65, 0x18, 0x09, 0x20, 0x01,
-	0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74,
-	0x6f, 0x62, 0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x08,
-	0x64, 0x65, 0x61, 0x64, 0x6c, 0x69, 0x6e, 0x65, 0x22, 0xad, 0x02, 0x0a, 0x0e, 0x53, 0x63, 0x68,
-	0x65, 0x64, 0x75, 0x6c, 0x65, 0x72, 0x45, 0x6e, 0x74, 0x72, 0x79, 0x12, 0x0e, 0x0a, 0x02, 0x69,
-	0x64, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x02, 0x69, 0x64, 0x12, 0x12, 0x0a, 0x04, 0x73,
-	0x70, 0x65, 0x63, 0x18, 0x02, 0x20, 0x01, 0x28, 0x09, 0x52, 0x04, 0x73, 0x70, 0x65, 0x63, 0x12,
-	0x1b, 0x0a, 0x09, 0x74, 0x61, 0x73, 0x6b, 0x5f, 0x74, 0x79, 0x70, 0x65, 0x18, 0x03, 0x20, 0x01,
-	0x28, 0x09, 0x52, 0x08, 0x74, 0x61, 0x73, 0x6b, 0x54, 0x79, 0x70, 0x65, 0x12, 0x21, 0x0a, 0x0c,
-	0x74, 0x61, 0x73, 0x6b, 0x5f, 0x70, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x18, 0x04, 0x20, 0x01,
-	0x28, 0x0c, 0x52, 0x0b, 0x74, 0x61, 0x73, 0x6b, 0x50, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x12,
-	0x27, 0x0a, 0x0f, 0x65, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x5f, 0x6f, 0x70, 0x74, 0x69, 0x6f,
-	0x6e, 0x73, 0x18, 0x05, 0x20, 0x03, 0x28, 0x09, 0x52, 0x0e, 0x65, 0x6e, 0x71, 0x75, 0x65, 0x75,
-	0x65, 0x4f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x12, 0x46, 0x0a, 0x11, 0x6e, 0x65, 0x78, 0x74,
-	0x5f, 0x65, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x5f, 0x74, 0x69, 0x6d, 0x65, 0x18, 0x06, 0x20,
-	0x01, 0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f,
-	0x74, 0x6f, 0x62, 0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52,
-	0x0f, 0x6e, 0x65, 0x78, 0x74, 0x45, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x54, 0x69, 0x6d, 0x65,
-	0x12, 0x46, 0x0a, 0x11, 0x70, 0x72, 0x65, 0x76, 0x5f, 0x65, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65,
-	0x5f, 0x74, 0x69, 0x6d, 0x65, 0x18, 0x07, 0x20, 0x01, 0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x67, 0x6f,
-	0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x75, 0x66, 0x2e, 0x54, 0x69,
-	0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x0f, 0x70, 0x72, 0x65, 0x76, 0x45, 0x6e, 0x71,
-	0x75, 0x65, 0x75, 0x65, 0x54, 0x69, 0x6d, 0x65, 0x22, 0x6f, 0x0a, 0x15, 0x53, 0x63, 0x68, 0x65,
-	0x64, 0x75, 0x6c, 0x65, 0x72, 0x45, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x45, 0x76, 0x65, 0x6e,
-	0x74, 0x12, 0x17, 0x0a, 0x07, 0x74, 0x61, 0x73, 0x6b, 0x5f, 0x69, 0x64, 0x18, 0x01, 0x20, 0x01,
-	0x28, 0x09, 0x52, 0x06, 0x74, 0x61, 0x73, 0x6b, 0x49, 0x64, 0x12, 0x3d, 0x0a, 0x0c, 0x65, 0x6e,
-	0x71, 0x75, 0x65, 0x75, 0x65, 0x5f, 0x74, 0x69, 0x6d, 0x65, 0x18, 0x02, 0x20, 0x01, 0x28, 0x0b,
+	0x79, 0x12, 0x1b, 0x0a, 0x09, 0x67, 0x72, 0x6f, 0x75, 0x70, 0x5f, 0x6b, 0x65, 0x79, 0x18, 0x0e,
+	0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x67, 0x72, 0x6f, 0x75, 0x70, 0x4b, 0x65, 0x79, 0x12, 0x1c,
+	0x0a, 0x09, 0x72, 0x65, 0x74, 0x65, 0x6e, 0x74, 0x69, 0x6f, 0x6e, 0x18, 0x0c, 0x20, 0x01, 0x28,
+	0x03, 0x52, 0x09, 0x72, 0x65, 0x74, 0x65, 0x6e, 0x74, 0x69, 0x6f, 0x6e, 0x12, 0x21, 0x0a, 0x0c,
+	0x63, 0x6f, 0x6d, 0x70, 0x6c, 0x65, 0x74, 0x65, 0x64, 0x5f, 0x61, 0x74, 0x18, 0x0d, 0x20, 0x01,
+	0x28, 0x03, 0x52, 0x0b, 0x63, 0x6f, 0x6d, 0x70, 0x6c, 0x65, 0x74, 0x65, 0x64, 0x41, 0x74, 0x22,
+	0x8f, 0x03, 0x0a, 0x0a, 0x53, 0x65, 0x72, 0x76, 0x65, 0x72, 0x49, 0x6e, 0x66, 0x6f, 0x12, 0x12,
+	0x0a, 0x04, 0x68, 0x6f, 0x73, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x04, 0x68, 0x6f,
+	0x73, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x70, 0x69, 0x64, 0x18, 0x02, 0x20, 0x01, 0x28, 0x05, 0x52,
+	0x03, 0x70, 0x69, 0x64, 0x12, 0x1b, 0x0a, 0x09, 0x73, 0x65, 0x72, 0x76, 0x65, 0x72, 0x5f, 0x69,
+	0x64, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x73, 0x65, 0x72, 0x76, 0x65, 0x72, 0x49,
+	0x64, 0x12, 0x20, 0x0a, 0x0b, 0x63, 0x6f, 0x6e, 0x63, 0x75, 0x72, 0x72, 0x65, 0x6e, 0x63, 0x79,
+	0x18, 0x04, 0x20, 0x01, 0x28, 0x05, 0x52, 0x0b, 0x63, 0x6f, 0x6e, 0x63, 0x75, 0x72, 0x72, 0x65,
+	0x6e, 0x63, 0x79, 0x12, 0x35, 0x0a, 0x06, 0x71, 0x75, 0x65, 0x75, 0x65, 0x73, 0x18, 0x05, 0x20,
+	0x03, 0x28, 0x0b, 0x32, 0x1d, 0x2e, 0x61, 0x73, 0x79, 0x6e, 0x71, 0x2e, 0x53, 0x65, 0x72, 0x76,
+	0x65, 0x72, 0x49, 0x6e, 0x66, 0x6f, 0x2e, 0x51, 0x75, 0x65, 0x75, 0x65, 0x73, 0x45, 0x6e, 0x74,
+	0x72, 0x79, 0x52, 0x06, 0x71, 0x75, 0x65, 0x75, 0x65, 0x73, 0x12, 0x27, 0x0a, 0x0f, 0x73, 0x74,
+	0x72, 0x69, 0x63, 0x74, 0x5f, 0x70, 0x72, 0x69, 0x6f, 0x72, 0x69, 0x74, 0x79, 0x18, 0x06, 0x20,
+	0x01, 0x28, 0x08, 0x52, 0x0e, 0x73, 0x74, 0x72, 0x69, 0x63, 0x74, 0x50, 0x72, 0x69, 0x6f, 0x72,
+	0x69, 0x74, 0x79, 0x12, 0x16, 0x0a, 0x06, 0x73, 0x74, 0x61, 0x74, 0x75, 0x73, 0x18, 0x07, 0x20,
+	0x01, 0x28, 0x09, 0x52, 0x06, 0x73, 0x74, 0x61, 0x74, 0x75, 0x73, 0x12, 0x39, 0x0a, 0x0a, 0x73,
+	0x74, 0x61, 0x72, 0x74, 0x5f, 0x74, 0x69, 0x6d, 0x65, 0x18, 0x08, 0x20, 0x01, 0x28, 0x0b, 0x32,
+	0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x75,
+	0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x09, 0x73, 0x74, 0x61,
+	0x72, 0x74, 0x54, 0x69, 0x6d, 0x65, 0x12, 0x2e, 0x0a, 0x13, 0x61, 0x63, 0x74, 0x69, 0x76, 0x65,
+	0x5f, 0x77, 0x6f, 0x72, 0x6b, 0x65, 0x72, 0x5f, 0x63, 0x6f, 0x75, 0x6e, 0x74, 0x18, 0x09, 0x20,
+	0x01, 0x28, 0x05, 0x52, 0x11, 0x61, 0x63, 0x74, 0x69, 0x76, 0x65, 0x57, 0x6f, 0x72, 0x6b, 0x65,
+	0x72, 0x43, 0x6f, 0x75, 0x6e, 0x74, 0x1a, 0x39, 0x0a, 0x0b, 0x51, 0x75, 0x65, 0x75, 0x65, 0x73,
+	0x45, 0x6e, 0x74, 0x72, 0x79, 0x12, 0x10, 0x0a, 0x03, 0x6b, 0x65, 0x79, 0x18, 0x01, 0x20, 0x01,
+	0x28, 0x09, 0x52, 0x03, 0x6b, 0x65, 0x79, 0x12, 0x14, 0x0a, 0x05, 0x76, 0x61, 0x6c, 0x75, 0x65,
+	0x18, 0x02, 0x20, 0x01, 0x28, 0x05, 0x52, 0x05, 0x76, 0x61, 0x6c, 0x75, 0x65, 0x3a, 0x02, 0x38,
+	0x01, 0x22, 0xb1, 0x02, 0x0a, 0x0a, 0x57, 0x6f, 0x72, 0x6b, 0x65, 0x72, 0x49, 0x6e, 0x66, 0x6f,
+	0x12, 0x12, 0x0a, 0x04, 0x68, 0x6f, 0x73, 0x74, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52, 0x04,
+	0x68, 0x6f, 0x73, 0x74, 0x12, 0x10, 0x0a, 0x03, 0x70, 0x69, 0x64, 0x18, 0x02, 0x20, 0x01, 0x28,
+	0x05, 0x52, 0x03, 0x70, 0x69, 0x64, 0x12, 0x1b, 0x0a, 0x09, 0x73, 0x65, 0x72, 0x76, 0x65, 0x72,
+	0x5f, 0x69, 0x64, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52, 0x08, 0x73, 0x65, 0x72, 0x76, 0x65,
+	0x72, 0x49, 0x64, 0x12, 0x17, 0x0a, 0x07, 0x74, 0x61, 0x73, 0x6b, 0x5f, 0x69, 0x64, 0x18, 0x04,
+	0x20, 0x01, 0x28, 0x09, 0x52, 0x06, 0x74, 0x61, 0x73, 0x6b, 0x49, 0x64, 0x12, 0x1b, 0x0a, 0x09,
+	0x74, 0x61, 0x73, 0x6b, 0x5f, 0x74, 0x79, 0x70, 0x65, 0x18, 0x05, 0x20, 0x01, 0x28, 0x09, 0x52,
+	0x08, 0x74, 0x61, 0x73, 0x6b, 0x54, 0x79, 0x70, 0x65, 0x12, 0x21, 0x0a, 0x0c, 0x74, 0x61, 0x73,
+	0x6b, 0x5f, 0x70, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x18, 0x06, 0x20, 0x01, 0x28, 0x0c, 0x52,
+	0x0b, 0x74, 0x61, 0x73, 0x6b, 0x50, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x12, 0x14, 0x0a, 0x05,
+	0x71, 0x75, 0x65, 0x75, 0x65, 0x18, 0x07, 0x20, 0x01, 0x28, 0x09, 0x52, 0x05, 0x71, 0x75, 0x65,
+	0x75, 0x65, 0x12, 0x39, 0x0a, 0x0a, 0x73, 0x74, 0x61, 0x72, 0x74, 0x5f, 0x74, 0x69, 0x6d, 0x65,
+	0x18, 0x08, 0x20, 0x01, 0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e,
+	0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61,
+	0x6d, 0x70, 0x52, 0x09, 0x73, 0x74, 0x61, 0x72, 0x74, 0x54, 0x69, 0x6d, 0x65, 0x12, 0x36, 0x0a,
+	0x08, 0x64, 0x65, 0x61, 0x64, 0x6c, 0x69, 0x6e, 0x65, 0x18, 0x09, 0x20, 0x01, 0x28, 0x0b, 0x32,
+	0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x75,
+	0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x08, 0x64, 0x65, 0x61,
+	0x64, 0x6c, 0x69, 0x6e, 0x65, 0x22, 0xad, 0x02, 0x0a, 0x0e, 0x53, 0x63, 0x68, 0x65, 0x64, 0x75,
+	0x6c, 0x65, 0x72, 0x45, 0x6e, 0x74, 0x72, 0x79, 0x12, 0x0e, 0x0a, 0x02, 0x69, 0x64, 0x18, 0x01,
+	0x20, 0x01, 0x28, 0x09, 0x52, 0x02, 0x69, 0x64, 0x12, 0x12, 0x0a, 0x04, 0x73, 0x70, 0x65, 0x63,
+	0x18, 0x02, 0x20, 0x01, 0x28, 0x09, 0x52, 0x04, 0x73, 0x70, 0x65, 0x63, 0x12, 0x1b, 0x0a, 0x09,
+	0x74, 0x61, 0x73, 0x6b, 0x5f, 0x74, 0x79, 0x70, 0x65, 0x18, 0x03, 0x20, 0x01, 0x28, 0x09, 0x52,
+	0x08, 0x74, 0x61, 0x73, 0x6b, 0x54, 0x79, 0x70, 0x65, 0x12, 0x21, 0x0a, 0x0c, 0x74, 0x61, 0x73,
+	0x6b, 0x5f, 0x70, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x18, 0x04, 0x20, 0x01, 0x28, 0x0c, 0x52,
+	0x0b, 0x74, 0x61, 0x73, 0x6b, 0x50, 0x61, 0x79, 0x6c, 0x6f, 0x61, 0x64, 0x12, 0x27, 0x0a, 0x0f,
+	0x65, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x5f, 0x6f, 0x70, 0x74, 0x69, 0x6f, 0x6e, 0x73, 0x18,
+	0x05, 0x20, 0x03, 0x28, 0x09, 0x52, 0x0e, 0x65, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x4f, 0x70,
+	0x74, 0x69, 0x6f, 0x6e, 0x73, 0x12, 0x46, 0x0a, 0x11, 0x6e, 0x65, 0x78, 0x74, 0x5f, 0x65, 0x6e,
+	0x71, 0x75, 0x65, 0x75, 0x65, 0x5f, 0x74, 0x69, 0x6d, 0x65, 0x18, 0x06, 0x20, 0x01, 0x28, 0x0b,
 	0x32, 0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62,
-	0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x0b, 0x65, 0x6e,
-	0x71, 0x75, 0x65, 0x75, 0x65, 0x54, 0x69, 0x6d, 0x65, 0x42, 0x29, 0x5a, 0x27, 0x67, 0x69, 0x74,
-	0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x68, 0x69, 0x62, 0x69, 0x6b, 0x65, 0x6e, 0x2f,
-	0x61, 0x73, 0x79, 0x6e, 0x71, 0x2f, 0x69, 0x6e, 0x74, 0x65, 0x72, 0x6e, 0x61, 0x6c, 0x2f, 0x70,
-	0x72, 0x6f, 0x74, 0x6f, 0x62, 0x06, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x33,
+	0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x0f, 0x6e, 0x65,
+	0x78, 0x74, 0x45, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x54, 0x69, 0x6d, 0x65, 0x12, 0x46, 0x0a,
+	0x11, 0x70, 0x72, 0x65, 0x76, 0x5f, 0x65, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x5f, 0x74, 0x69,
+	0x6d, 0x65, 0x18, 0x07, 0x20, 0x01, 0x28, 0x0b, 0x32, 0x1a, 0x2e, 0x67, 0x6f, 0x6f, 0x67, 0x6c,
+	0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x75, 0x66, 0x2e, 0x54, 0x69, 0x6d, 0x65, 0x73,
+	0x74, 0x61, 0x6d, 0x70, 0x52, 0x0f, 0x70, 0x72, 0x65, 0x76, 0x45, 0x6e, 0x71, 0x75, 0x65, 0x75,
+	0x65, 0x54, 0x69, 0x6d, 0x65, 0x22, 0x6f, 0x0a, 0x15, 0x53, 0x63, 0x68, 0x65, 0x64, 0x75, 0x6c,
+	0x65, 0x72, 0x45, 0x6e, 0x71, 0x75, 0x65, 0x75, 0x65, 0x45, 0x76, 0x65, 0x6e, 0x74, 0x12, 0x17,
+	0x0a, 0x07, 0x74, 0x61, 0x73, 0x6b, 0x5f, 0x69, 0x64, 0x18, 0x01, 0x20, 0x01, 0x28, 0x09, 0x52,
+	0x06, 0x74, 0x61, 0x73, 0x6b, 0x49, 0x64, 0x12, 0x3d, 0x0a, 0x0c, 0x65, 0x6e, 0x71, 0x75, 0x65,
+	0x75, 0x65, 0x5f, 0x74, 0x69, 0x6d, 0x65, 0x18, 0x02, 0x20, 0x01, 0x28, 0x0b, 0x32, 0x1a, 0x2e,
+	0x67, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x2e, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x62, 0x75, 0x66, 0x2e,
+	0x54, 0x69, 0x6d, 0x65, 0x73, 0x74, 0x61, 0x6d, 0x70, 0x52, 0x0b, 0x65, 0x6e, 0x71, 0x75, 0x65,
+	0x75, 0x65, 0x54, 0x69, 0x6d, 0x65, 0x42, 0x29, 0x5a, 0x27, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62,
+	0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x68, 0x69, 0x62, 0x69, 0x6b, 0x65, 0x6e, 0x2f, 0x61, 0x73, 0x79,
+	0x6e, 0x71, 0x2f, 0x69, 0x6e, 0x74, 0x65, 0x72, 0x6e, 0x61, 0x6c, 0x2f, 0x70, 0x72, 0x6f, 0x74,
+	0x6f, 0x62, 0x06, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x33,
 }
 
 var (

internal/proto/asynq.proto

@@ -11,6 +11,7 @@ option go_package = "github.com/hibiken/asynq/internal/proto";
 
 // TaskMessage is the internal representation of a task with additional
 // metadata fields.
+// Next ID: 15
 message TaskMessage {
 	// Type indicates the kind of the task to be performed.
   string type = 1;
@@ -51,6 +52,10 @@ message TaskMessage {
 	// Empty string indicates that no uniqueness lock was used.
   string unique_key = 10;
 
+  // GroupKey is a name of the group used for task aggregation.
+  // This field is optional and empty value means no aggregation for the task.
+  string group_key = 14;
+
   // Retention period specified in a number of seconds.
   // The task will be stored in redis as a completed task until the TTL
   // expires.

internal/rdb/benchmark_test.go

@@ -10,19 +10,19 @@ import (
 	"testing"
 	"time"
 
-	"github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
+	"github.com/hibiken/asynq/internal/testutil"
 )
 
 func BenchmarkEnqueue(b *testing.B) {
 	r := setup(b)
 	ctx := context.Background()
-	msg := asynqtest.NewTaskMessage("task1", nil)
+	msg := testutil.NewTaskMessage("task1", nil)
 	b.ResetTimer()
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
+		testutil.FlushDB(b, r.client)
 		b.StartTimer()
 
 		if err := r.Enqueue(ctx, msg); err != nil {
@@ -45,7 +45,7 @@ func BenchmarkEnqueueUnique(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
+		testutil.FlushDB(b, r.client)
 		b.StartTimer()
 
 		if err := r.EnqueueUnique(ctx, msg, uniqueTTL); err != nil {
@@ -57,13 +57,13 @@ func BenchmarkEnqueueUnique(b *testing.B) {
 func BenchmarkSchedule(b *testing.B) {
 	r := setup(b)
 	ctx := context.Background()
-	msg := asynqtest.NewTaskMessage("task1", nil)
+	msg := testutil.NewTaskMessage("task1", nil)
 	processAt := time.Now().Add(3 * time.Minute)
 	b.ResetTimer()
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
+		testutil.FlushDB(b, r.client)
 		b.StartTimer()
 
 		if err := r.Schedule(ctx, msg, processAt); err != nil {
@@ -87,7 +87,7 @@ func BenchmarkScheduleUnique(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
+		testutil.FlushDB(b, r.client)
 		b.StartTimer()
 
 		if err := r.ScheduleUnique(ctx, msg, processAt, uniqueTTL); err != nil {
@@ -103,9 +103,9 @@ func BenchmarkDequeueSingleQueue(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
+		testutil.FlushDB(b, r.client)
 		for i := 0; i < 10; i++ {
-			m := asynqtest.NewTaskMessageWithQueue(
+			m := testutil.NewTaskMessageWithQueue(
 				fmt.Sprintf("task%d", i), nil, base.DefaultQueueName)
 			if err := r.Enqueue(ctx, m); err != nil {
 				b.Fatalf("Enqueue failed: %v", err)
@@ -127,10 +127,10 @@ func BenchmarkDequeueMultipleQueues(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
+		testutil.FlushDB(b, r.client)
 		for i := 0; i < 10; i++ {
 			for _, qname := range qnames {
-				m := asynqtest.NewTaskMessageWithQueue(
+				m := testutil.NewTaskMessageWithQueue(
 					fmt.Sprintf("%s_task%d", qname, i), nil, qname)
 				if err := r.Enqueue(ctx, m); err != nil {
 					b.Fatalf("Enqueue failed: %v", err)
@@ -147,9 +147,9 @@ func BenchmarkDequeueMultipleQueues(b *testing.B) {
 
 func BenchmarkDone(b *testing.B) {
 	r := setup(b)
-	m1 := asynqtest.NewTaskMessage("task1", nil)
-	m2 := asynqtest.NewTaskMessage("task2", nil)
-	m3 := asynqtest.NewTaskMessage("task3", nil)
+	m1 := testutil.NewTaskMessage("task1", nil)
+	m2 := testutil.NewTaskMessage("task2", nil)
+	m3 := testutil.NewTaskMessage("task3", nil)
 	msgs := []*base.TaskMessage{m1, m2, m3}
 	zs := []base.Z{
 		{Message: m1, Score: time.Now().Add(10 * time.Second).Unix()},
@@ -161,9 +161,9 @@ func BenchmarkDone(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
-		asynqtest.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
-		asynqtest.SeedLease(b, r.client, zs, base.DefaultQueueName)
+		testutil.FlushDB(b, r.client)
+		testutil.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
+		testutil.SeedLease(b, r.client, zs, base.DefaultQueueName)
 		b.StartTimer()
 
 		if err := r.Done(ctx, msgs[0]); err != nil {
@@ -174,9 +174,9 @@ func BenchmarkDone(b *testing.B) {
 
 func BenchmarkRetry(b *testing.B) {
 	r := setup(b)
-	m1 := asynqtest.NewTaskMessage("task1", nil)
-	m2 := asynqtest.NewTaskMessage("task2", nil)
-	m3 := asynqtest.NewTaskMessage("task3", nil)
+	m1 := testutil.NewTaskMessage("task1", nil)
+	m2 := testutil.NewTaskMessage("task2", nil)
+	m3 := testutil.NewTaskMessage("task3", nil)
 	msgs := []*base.TaskMessage{m1, m2, m3}
 	zs := []base.Z{
 		{Message: m1, Score: time.Now().Add(10 * time.Second).Unix()},
@@ -188,9 +188,9 @@ func BenchmarkRetry(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
-		asynqtest.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
-		asynqtest.SeedLease(b, r.client, zs, base.DefaultQueueName)
+		testutil.FlushDB(b, r.client)
+		testutil.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
+		testutil.SeedLease(b, r.client, zs, base.DefaultQueueName)
 		b.StartTimer()
 
 		if err := r.Retry(ctx, msgs[0], time.Now().Add(1*time.Minute), "error", true /*isFailure*/); err != nil {
@@ -201,9 +201,9 @@ func BenchmarkRetry(b *testing.B) {
 
 func BenchmarkArchive(b *testing.B) {
 	r := setup(b)
-	m1 := asynqtest.NewTaskMessage("task1", nil)
-	m2 := asynqtest.NewTaskMessage("task2", nil)
-	m3 := asynqtest.NewTaskMessage("task3", nil)
+	m1 := testutil.NewTaskMessage("task1", nil)
+	m2 := testutil.NewTaskMessage("task2", nil)
+	m3 := testutil.NewTaskMessage("task3", nil)
 	msgs := []*base.TaskMessage{m1, m2, m3}
 	zs := []base.Z{
 		{Message: m1, Score: time.Now().Add(10 * time.Second).Unix()},
@@ -215,9 +215,9 @@ func BenchmarkArchive(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
-		asynqtest.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
-		asynqtest.SeedLease(b, r.client, zs, base.DefaultQueueName)
+		testutil.FlushDB(b, r.client)
+		testutil.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
+		testutil.SeedLease(b, r.client, zs, base.DefaultQueueName)
 		b.StartTimer()
 
 		if err := r.Archive(ctx, msgs[0], "error"); err != nil {
@@ -228,9 +228,9 @@ func BenchmarkArchive(b *testing.B) {
 
 func BenchmarkRequeue(b *testing.B) {
 	r := setup(b)
-	m1 := asynqtest.NewTaskMessage("task1", nil)
-	m2 := asynqtest.NewTaskMessage("task2", nil)
-	m3 := asynqtest.NewTaskMessage("task3", nil)
+	m1 := testutil.NewTaskMessage("task1", nil)
+	m2 := testutil.NewTaskMessage("task2", nil)
+	m3 := testutil.NewTaskMessage("task3", nil)
 	msgs := []*base.TaskMessage{m1, m2, m3}
 	zs := []base.Z{
 		{Message: m1, Score: time.Now().Add(10 * time.Second).Unix()},
@@ -242,9 +242,9 @@ func BenchmarkRequeue(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
-		asynqtest.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
-		asynqtest.SeedLease(b, r.client, zs, base.DefaultQueueName)
+		testutil.FlushDB(b, r.client)
+		testutil.SeedActiveQueue(b, r.client, msgs, base.DefaultQueueName)
+		testutil.SeedLease(b, r.client, zs, base.DefaultQueueName)
 		b.StartTimer()
 
 		if err := r.Requeue(ctx, msgs[0]); err != nil {
@@ -258,7 +258,7 @@ func BenchmarkCheckAndEnqueue(b *testing.B) {
 	now := time.Now()
 	var zs []base.Z
 	for i := -100; i < 100; i++ {
-		msg := asynqtest.NewTaskMessage(fmt.Sprintf("task%d", i), nil)
+		msg := testutil.NewTaskMessage(fmt.Sprintf("task%d", i), nil)
 		score := now.Add(time.Duration(i) * time.Second).Unix()
 		zs = append(zs, base.Z{Message: msg, Score: score})
 	}
@@ -266,8 +266,8 @@ func BenchmarkCheckAndEnqueue(b *testing.B) {
 
 	for i := 0; i < b.N; i++ {
 		b.StopTimer()
-		asynqtest.FlushDB(b, r.client)
-		asynqtest.SeedScheduledQueue(b, r.client, zs, base.DefaultQueueName)
+		testutil.FlushDB(b, r.client)
+		testutil.SeedScheduledQueue(b, r.client, zs, base.DefaultQueueName)
 		b.StartTimer()
 
 		if err := r.ForwardIfReady(base.DefaultQueueName); err != nil {

internal/rdb/inspect.go

@@ -34,13 +34,18 @@ type Stats struct {
 	Paused bool
 	// Size is the total number of tasks in the queue.
 	Size int
+
+	// Groups is the total number of groups in the queue.
+	Groups int
+
 	// Number of tasks in each state.
-	Pending   int
-	Active    int
-	Scheduled int
-	Retry     int
-	Archived  int
-	Completed int
+	Pending     int
+	Active      int
+	Scheduled   int
+	Retry       int
+	Archived    int
+	Completed   int
+	Aggregating int
 
 	// Number of tasks processed within the current date.
 	// The number includes both succeeded and failed tasks.
@@ -83,8 +88,10 @@ type DailyStats struct {
 // KEYS[9] ->  asynq:<qname>:processed
 // KEYS[10] -> asynq:<qname>:failed
 // KEYS[11] -> asynq:<qname>:paused
-//
+// KEYS[12] -> asynq:<qname>:groups
+// --------
 // ARGV[1] -> task key prefix
+// ARGV[2] -> group key prefix
 var currentStatsCmd = redis.NewScript(`
 local res = {}
 local pendingTaskCount = redis.call("LLEN", KEYS[1])
@@ -118,6 +125,15 @@ if pendingTaskCount > 0 then
 else
 	table.insert(res, 0)
 end
+local group_names = redis.call("SMEMBERS", KEYS[12])
+table.insert(res, "group_size")
+table.insert(res, table.getn(group_names))
+local aggregating_count = 0
+for _, gname in ipairs(group_names) do
+	aggregating_count = aggregating_count + redis.call("ZCARD", ARGV[2] .. gname)
+end
+table.insert(res, "aggregating_count")
+table.insert(res, aggregating_count)
 return res`)
 
 // CurrentStats returns a current state of the queues.
@@ -131,7 +147,7 @@ func (r *RDB) CurrentStats(qname string) (*Stats, error) {
 		return nil, errors.E(op, errors.NotFound, &errors.QueueNotFoundError{Queue: qname})
 	}
 	now := r.clock.Now()
-	res, err := currentStatsCmd.Run(context.Background(), r.client, []string{
+	keys := []string{
 		base.PendingKey(qname),
 		base.ActiveKey(qname),
 		base.ScheduledKey(qname),
@@ -143,7 +159,13 @@ func (r *RDB) CurrentStats(qname string) (*Stats, error) {
 		base.ProcessedTotalKey(qname),
 		base.FailedTotalKey(qname),
 		base.PausedKey(qname),
-	}, base.TaskKeyPrefix(qname)).Result()
+		base.AllGroups(qname),
+	}
+	argv := []interface{}{
+		base.TaskKeyPrefix(qname),
+		base.GroupKeyPrefix(qname),
+	}
+	res, err := currentStatsCmd.Run(context.Background(), r.client, keys, argv...).Result()
 	if err != nil {
 		return nil, errors.E(op, errors.Unknown, err)
 	}
@@ -198,6 +220,11 @@ func (r *RDB) CurrentStats(qname string) (*Stats, error) {
 			} else {
 				stats.Latency = r.clock.Now().Sub(time.Unix(0, int64(val)))
 			}
+		case "group_size":
+			stats.Groups = val
+		case "aggregating_count":
+			stats.Aggregating = val
+			size += val
 		}
 	}
 	stats.Size = size
@@ -219,9 +246,12 @@ func (r *RDB) CurrentStats(qname string) (*Stats, error) {
 // KEYS[4] -> asynq:{qname}:retry
 // KEYS[5] -> asynq:{qname}:archived
 // KEYS[6] -> asynq:{qname}:completed
-//
-// ARGV[1] -> asynq:{qname}:t:
-// ARGV[2] -> sample_size (e.g 20)
+// KEYS[7] -> asynq:{qname}:groups
+// -------
+// ARGV[1] -> asynq:{qname}:t: (task key prefix)
+// ARGV[2] -> task sample size per redis list/zset (e.g 20)
+// ARGV[3] -> group sample size
+// ARGV[4] -> asynq:{qname}:g: (group key prefix)
 var memoryUsageCmd = redis.NewScript(`
 local sample_size = tonumber(ARGV[2])
 if sample_size <= 0 then
@@ -262,12 +292,36 @@ for i=3,6 do
         memusg = memusg + m
     end
 end
+local groups = redis.call("SMEMBERS", KEYS[7])
+if table.getn(groups) > 0 then
+	local agg_task_count = 0
+	local agg_task_sample_total = 0
+	local agg_task_sample_size = 0
+	for i, gname in ipairs(groups) do
+		local group_key = ARGV[4] .. gname
+		agg_task_count = agg_task_count + redis.call("ZCARD", group_key)
+		if i <= tonumber(ARGV[3]) then
+			local ids = redis.call("ZRANGE", group_key, 0, sample_size - 1)
+			for _, id in ipairs(ids) do
+				local bytes = redis.call("MEMORY", "USAGE", ARGV[1] .. id)
+				agg_task_sample_total = agg_task_sample_total + bytes
+				agg_task_sample_size = agg_task_sample_size + 1
+			end
+		end
+	end
+	local avg = agg_task_sample_total / agg_task_sample_size
+	memusg = memusg + (avg * agg_task_count)
+end
 return memusg
 `)
 
 func (r *RDB) memoryUsage(qname string) (int64, error) {
 	var op errors.Op = "rdb.memoryUsage"
-	const sampleSize = 20
+	const (
+		taskSampleSize  = 20
+		groupSampleSize = 5
+	)
+
 	keys := []string{
 		base.ActiveKey(qname),
 		base.PendingKey(qname),
@@ -275,10 +329,13 @@ func (r *RDB) memoryUsage(qname string) (int64, error) {
 		base.RetryKey(qname),
 		base.ArchivedKey(qname),
 		base.CompletedKey(qname),
+		base.AllGroups(qname),
 	}
 	argv := []interface{}{
 		base.TaskKeyPrefix(qname),
-		sampleSize,
+		taskSampleSize,
+		groupSampleSize,
+		base.GroupKeyPrefix(qname),
 	}
 	res, err := memoryUsageCmd.Run(context.Background(), r.client, keys, argv...).Result()
 	if err != nil {
@@ -490,6 +547,56 @@ func (r *RDB) GetTaskInfo(qname, id string) (*base.TaskInfo, error) {
 	}, nil
 }
 
+type GroupStat struct {
+	// Name of the group.
+	Group string
+
+	// Size of the group.
+	Size int
+}
+
+// KEYS[1] -> asynq:{<qname>}:groups
+// -------
+// ARGV[1] -> group key prefix
+//
+// Output:
+// list of group name and size (e.g. group1 size1 group2 size2 ...)
+//
+// Time Complexity:
+// O(N) where N being the number of groups in the given queue.
+var groupStatsCmd = redis.NewScript(`
+local res = {}
+local group_names = redis.call("SMEMBERS", KEYS[1])
+for _, gname in ipairs(group_names) do
+	local size = redis.call("ZCARD", ARGV[1] .. gname) 
+	table.insert(res, gname)
+	table.insert(res, size)
+end
+return res
+`)
+
+func (r *RDB) GroupStats(qname string) ([]*GroupStat, error) {
+	var op errors.Op = "RDB.GroupStats"
+	keys := []string{base.AllGroups(qname)}
+	argv := []interface{}{base.GroupKeyPrefix(qname)}
+	res, err := groupStatsCmd.Run(context.Background(), r.client, keys, argv...).Result()
+	if err != nil {
+		return nil, errors.E(op, errors.Unknown, err)
+	}
+	data, err := cast.ToSliceE(res)
+	if err != nil {
+		return nil, errors.E(op, errors.Internal, "cast error: unexpected return value from Lua script")
+	}
+	var stats []*GroupStat
+	for i := 0; i < len(data); i += 2 {
+		stats = append(stats, &GroupStat{
+			Group: data[i].(string),
+			Size:  int(data[i+1].(int64)),
+		})
+	}
+	return stats, nil
+}
+
 // Pagination specifies the page size and page number
 // for the list operation.
 type Pagination struct {
@@ -619,7 +726,7 @@ func (r *RDB) ListScheduled(qname string, pgn Pagination) ([]*base.TaskInfo, err
 	if !exists {
 		return nil, errors.E(op, errors.NotFound, &errors.QueueNotFoundError{Queue: qname})
 	}
-	res, err := r.listZSetEntries(qname, base.TaskStateScheduled, pgn)
+	res, err := r.listZSetEntries(qname, base.TaskStateScheduled, base.ScheduledKey(qname), pgn)
 	if err != nil {
 		return nil, errors.E(op, errors.CanonicalCode(err), err)
 	}
@@ -637,7 +744,7 @@ func (r *RDB) ListRetry(qname string, pgn Pagination) ([]*base.TaskInfo, error)
 	if !exists {
 		return nil, errors.E(op, errors.NotFound, &errors.QueueNotFoundError{Queue: qname})
 	}
-	res, err := r.listZSetEntries(qname, base.TaskStateRetry, pgn)
+	res, err := r.listZSetEntries(qname, base.TaskStateRetry, base.RetryKey(qname), pgn)
 	if err != nil {
 		return nil, errors.E(op, errors.CanonicalCode(err), err)
 	}
@@ -654,7 +761,7 @@ func (r *RDB) ListArchived(qname string, pgn Pagination) ([]*base.TaskInfo, erro
 	if !exists {
 		return nil, errors.E(op, errors.NotFound, &errors.QueueNotFoundError{Queue: qname})
 	}
-	zs, err := r.listZSetEntries(qname, base.TaskStateArchived, pgn)
+	zs, err := r.listZSetEntries(qname, base.TaskStateArchived, base.ArchivedKey(qname), pgn)
 	if err != nil {
 		return nil, errors.E(op, errors.CanonicalCode(err), err)
 	}
@@ -671,7 +778,24 @@ func (r *RDB) ListCompleted(qname string, pgn Pagination) ([]*base.TaskInfo, err
 	if !exists {
 		return nil, errors.E(op, errors.NotFound, &errors.QueueNotFoundError{Queue: qname})
 	}
-	zs, err := r.listZSetEntries(qname, base.TaskStateCompleted, pgn)
+	zs, err := r.listZSetEntries(qname, base.TaskStateCompleted, base.CompletedKey(qname), pgn)
+	if err != nil {
+		return nil, errors.E(op, errors.CanonicalCode(err), err)
+	}
+	return zs, nil
+}
+
+// ListAggregating returns all tasks from the given group.
+func (r *RDB) ListAggregating(qname, gname string, pgn Pagination) ([]*base.TaskInfo, error) {
+	var op errors.Op = "rdb.ListAggregating"
+	exists, err := r.queueExists(qname)
+	if err != nil {
+		return nil, errors.E(op, errors.Unknown, &errors.RedisCommandError{Command: "sismember", Err: err})
+	}
+	if !exists {
+		return nil, errors.E(op, errors.NotFound, &errors.QueueNotFoundError{Queue: qname})
+	}
+	zs, err := r.listZSetEntries(qname, base.TaskStateAggregating, base.GroupKey(qname, gname), pgn)
 	if err != nil {
 		return nil, errors.E(op, errors.CanonicalCode(err), err)
 	}
@@ -707,20 +831,7 @@ return data
 
 // listZSetEntries returns a list of message and score pairs in Redis sorted-set
 // with the given key.
-func (r *RDB) listZSetEntries(qname string, state base.TaskState, pgn Pagination) ([]*base.TaskInfo, error) {
-	var key string
-	switch state {
-	case base.TaskStateScheduled:
-		key = base.ScheduledKey(qname)
-	case base.TaskStateRetry:
-		key = base.RetryKey(qname)
-	case base.TaskStateArchived:
-		key = base.ArchivedKey(qname)
-	case base.TaskStateCompleted:
-		key = base.CompletedKey(qname)
-	default:
-		panic(fmt.Sprintf("unsupported task state: %v", state))
-	}
+func (r *RDB) listZSetEntries(qname string, state base.TaskState, key string, pgn Pagination) ([]*base.TaskInfo, error) {
 	res, err := listZSetEntriesCmd.Run(context.Background(), r.client, []string{key},
 		pgn.start(), pgn.stop(), base.TaskKeyPrefix(qname)).Result()
 	if err != nil {
@@ -811,14 +922,67 @@ func (r *RDB) RunAllArchivedTasks(qname string) (int64, error) {
 	return n, nil
 }
 
+// runAllAggregatingCmd schedules all tasks in the group to run individually.
+//
+// Input:
+// KEYS[1] -> asynq:{<qname>}:g:<gname>
+// KEYS[2] -> asynq:{<qname>}:pending
+// KEYS[3] -> asynq:{<qname>}:groups
+// -------
+// ARGV[1] -> task key prefix
+// ARGV[2] -> group name
+//
+// Output:
+// integer: number of tasks scheduled to run
+var runAllAggregatingCmd = redis.NewScript(`
+local ids = redis.call("ZRANGE", KEYS[1], 0, -1)
+for _, id in ipairs(ids) do
+	redis.call("LPUSH", KEYS[2], id)
+	redis.call("HSET", ARGV[1] .. id, "state", "pending")
+end
+redis.call("DEL", KEYS[1])
+redis.call("SREM", KEYS[3], ARGV[2])
+return table.getn(ids)
+`)
+
+// RunAllAggregatingTasks schedules all tasks from the given queue to run
+// and returns the number of tasks scheduled to run.
+// If a queue with the given name doesn't exist, it returns QueueNotFoundError.
+func (r *RDB) RunAllAggregatingTasks(qname, gname string) (int64, error) {
+	var op errors.Op = "rdb.RunAllAggregatingTasks"
+	if err := r.checkQueueExists(qname); err != nil {
+		return 0, errors.E(op, errors.CanonicalCode(err), err)
+	}
+	keys := []string{
+		base.GroupKey(qname, gname),
+		base.PendingKey(qname),
+		base.AllGroups(qname),
+	}
+	argv := []interface{}{
+		base.TaskKeyPrefix(qname),
+		gname,
+	}
+	res, err := runAllAggregatingCmd.Run(context.Background(), r.client, keys, argv...).Result()
+	if err != nil {
+		return 0, errors.E(op, errors.Internal, err)
+	}
+	n, ok := res.(int64)
+	if !ok {
+		return 0, errors.E(op, errors.Internal, fmt.Sprintf("unexpected return value from script %v", res))
+	}
+	return n, nil
+}
+
 // runTaskCmd is a Lua script that updates the given task to pending state.
 //
 // Input:
 // KEYS[1] -> asynq:{<qname>}:t:<task_id>
 // KEYS[2] -> asynq:{<qname>}:pending
+// KEYS[3] -> asynq:{<qname>}:groups
 // --
 // ARGV[1] -> task ID
 // ARGV[2] -> queue key prefix; asynq:{<qname>}:
+// ARGV[3] -> group key prefix
 //
 // Output:
 // Numeric code indicating the status:
@@ -831,15 +995,24 @@ var runTaskCmd = redis.NewScript(`
 if redis.call("EXISTS", KEYS[1]) == 0 then
 	return 0
 end
-local state = redis.call("HGET", KEYS[1], "state")
+local state, group = unpack(redis.call("HMGET", KEYS[1], "state", "group"))
 if state == "active" then
 	return -1
 elseif state == "pending" then
 	return -2
-end
-local n = redis.call("ZREM", ARGV[2] .. state, ARGV[1])
-if n == 0 then
-	return redis.error_reply("internal error: task id not found in zset " .. tostring(state))
+elseif state == "aggregating" then
+	local n = redis.call("ZREM", ARGV[3] .. group, ARGV[1])
+	if n == 0 then
+		return redis.error_reply("internal error: task id not found in zset " .. tostring(ARGV[3] .. group))
+	end
+	if redis.call("ZCARD", ARGV[3] .. group) == 0 then
+		redis.call("SREM", KEYS[3], group)
+	end
+else
+	local n = redis.call("ZREM", ARGV[2] .. state, ARGV[1])
+	if n == 0 then
+		return redis.error_reply("internal error: task id not found in zset " .. tostring(ARGV[2] .. state))
+	end
 end
 redis.call("LPUSH", KEYS[2], ARGV[1])
 redis.call("HSET", KEYS[1], "state", "pending")
@@ -860,10 +1033,12 @@ func (r *RDB) RunTask(qname, id string) error {
 	keys := []string{
 		base.TaskKey(qname, id),
 		base.PendingKey(qname),
+		base.AllGroups(qname),
 	}
 	argv := []interface{}{
 		id,
 		base.QueueKeyPrefix(qname),
+		base.GroupKeyPrefix(qname),
 	}
 	res, err := runTaskCmd.Run(context.Background(), r.client, keys, argv...).Result()
 	if err != nil {
@@ -962,6 +1137,66 @@ func (r *RDB) ArchiveAllScheduledTasks(qname string) (int64, error) {
 	return n, nil
 }
 
+// archiveAllAggregatingCmd archives all tasks in the given group.
+//
+// Input:
+// KEYS[1] -> asynq:{<qname>}:g:<gname>
+// KEYS[2] -> asynq:{<qname>}:archived
+// KEYS[3] -> asynq:{<qname>}:groups
+// -------
+// ARGV[1] -> current timestamp
+// ARGV[2] -> cutoff timestamp (e.g., 90 days ago)
+// ARGV[3] -> max number of tasks in archive (e.g., 100)
+// ARGV[4] -> task key prefix (asynq:{<qname>}:t:)
+// ARGV[5] -> group name
+//
+// Output:
+// integer: Number of tasks archived
+var archiveAllAggregatingCmd = redis.NewScript(`
+local ids = redis.call("ZRANGE", KEYS[1], 0, -1)
+for _, id in ipairs(ids) do
+	redis.call("ZADD", KEYS[2], ARGV[1], id)
+	redis.call("HSET", ARGV[4] .. id, "state", "archived")
+end
+redis.call("ZREMRANGEBYSCORE", KEYS[2], "-inf", ARGV[2])
+redis.call("ZREMRANGEBYRANK", KEYS[2], 0, -ARGV[3])
+redis.call("DEL", KEYS[1])
+redis.call("SREM", KEYS[3], ARGV[5])
+return table.getn(ids)
+`)
+
+// ArchiveAllAggregatingTasks archives all aggregating tasks from the given group
+// and returns the number of tasks archived.
+// If a queue with the given name doesn't exist, it returns QueueNotFoundError.
+func (r *RDB) ArchiveAllAggregatingTasks(qname, gname string) (int64, error) {
+	var op errors.Op = "rdb.ArchiveAllAggregatingTasks"
+	if err := r.checkQueueExists(qname); err != nil {
+		return 0, errors.E(op, errors.CanonicalCode(err), err)
+	}
+	keys := []string{
+		base.GroupKey(qname, gname),
+		base.ArchivedKey(qname),
+		base.AllGroups(qname),
+	}
+	now := r.clock.Now()
+	argv := []interface{}{
+		now.Unix(),
+		now.AddDate(0, 0, -archivedExpirationInDays).Unix(),
+		maxArchiveSize,
+		base.TaskKeyPrefix(qname),
+		gname,
+	}
+	res, err := archiveAllAggregatingCmd.Run(context.Background(), r.client, keys, argv...).Result()
+	if err != nil {
+		return 0, errors.E(op, errors.Internal, err)
+	}
+	n, ok := res.(int64)
+	if !ok {
+		return 0, errors.E(op, errors.Internal, fmt.Sprintf("unexpected return value from script %v", res))
+	}
+	return n, nil
+}
+
 // archiveAllPendingCmd is a Lua script that moves all pending tasks from
 // the given queue to archived state.
 //
@@ -1022,12 +1257,14 @@ func (r *RDB) ArchiveAllPendingTasks(qname string) (int64, error) {
 // Input:
 // KEYS[1] -> task key (asynq:{<qname>}:t:<task_id>)
 // KEYS[2] -> archived key (asynq:{<qname>}:archived)
+// KEYS[3] -> all groups key (asynq:{<qname>}:groups)
 // --
 // ARGV[1] -> id of the task to archive
 // ARGV[2] -> current timestamp
 // ARGV[3] -> cutoff timestamp (e.g., 90 days ago)
 // ARGV[4] -> max number of tasks in archived state (e.g., 100)
 // ARGV[5] -> queue key prefix (asynq:{<qname>}:)
+// ARGV[6] -> group key prefix (asynq:{<qname>}:g:)
 //
 // Output:
 // Numeric code indicating the status:
@@ -1040,7 +1277,7 @@ var archiveTaskCmd = redis.NewScript(`
 if redis.call("EXISTS", KEYS[1]) == 0 then
 	return 0
 end
-local state = redis.call("HGET", KEYS[1], "state")
+local state, group = unpack(redis.call("HMGET", KEYS[1], "state", "group"))
 if state == "active" then
 	return -2
 end
@@ -1049,11 +1286,18 @@ if state == "archived" then
 end
 if state == "pending" then
 	if redis.call("LREM", ARGV[5] .. state, 1, ARGV[1]) == 0 then
-		return redis.error_reply("task id not found in list " .. tostring(state))
+		return redis.error_reply("task id not found in list " .. tostring(ARGV[5] .. state))
+	end
+elseif state == "aggregating" then
+	if redis.call("ZREM", ARGV[6] .. group, ARGV[1]) == 0 then
+		return redis.error_reply("task id not found in zset " .. tostring(ARGV[6] .. group))
+	end
+	if redis.call("ZCARD", ARGV[6] .. group) == 0 then
+		redis.call("SREM", KEYS[3], group)
 	end
 else 
 	if redis.call("ZREM", ARGV[5] .. state, ARGV[1]) == 0 then
-		return redis.error_reply("task id not found in zset " .. tostring(state))
+		return redis.error_reply("task id not found in zset " .. tostring(ARGV[5] .. state))
 	end
 end
 redis.call("ZADD", KEYS[2], ARGV[2], ARGV[1])
@@ -1078,6 +1322,7 @@ func (r *RDB) ArchiveTask(qname, id string) error {
 	keys := []string{
 		base.TaskKey(qname, id),
 		base.ArchivedKey(qname),
+		base.AllGroups(qname),
 	}
 	now := r.clock.Now()
 	argv := []interface{}{
@@ -1086,6 +1331,7 @@ func (r *RDB) ArchiveTask(qname, id string) error {
 		now.AddDate(0, 0, -archivedExpirationInDays).Unix(),
 		maxArchiveSize,
 		base.QueueKeyPrefix(qname),
+		base.GroupKeyPrefix(qname),
 	}
 	res, err := archiveTaskCmd.Run(context.Background(), r.client, keys, argv...).Result()
 	if err != nil {
@@ -1168,9 +1414,11 @@ func (r *RDB) archiveAll(src, dst, qname string) (int64, error) {
 
 // Input:
 // KEYS[1] -> asynq:{<qname>}:t:<task_id>
+// KEYS[2] -> asynq:{<qname>}:groups
 // --
 // ARGV[1] -> task ID
 // ARGV[2] -> queue key prefix
+// ARGV[3] -> group key prefix
 //
 // Output:
 // Numeric code indicating the status:
@@ -1181,17 +1429,24 @@ var deleteTaskCmd = redis.NewScript(`
 if redis.call("EXISTS", KEYS[1]) == 0 then
 	return 0
 end
-local state = redis.call("HGET", KEYS[1], "state")
+local state, group = unpack(redis.call("HMGET", KEYS[1], "state", "group"))
 if state == "active" then
 	return -1
 end
 if state == "pending" then
 	if redis.call("LREM", ARGV[2] .. state, 0, ARGV[1]) == 0 then
-		return redis.error_reply("task is not found in list: " .. tostring(state))
+		return redis.error_reply("task is not found in list: " .. tostring(ARGV[2] .. state))
+	end
+elseif state == "aggregating" then
+	if redis.call("ZREM", ARGV[3] .. group, ARGV[1]) == 0 then
+		return redis.error_reply("task is not found in zset: " .. tostring(ARGV[3] .. group))
+	end
+	if redis.call("ZCARD", ARGV[3] .. group) == 0 then
+		redis.call("SREM", KEYS[2], group)
 	end
 else
 	if redis.call("ZREM", ARGV[2] .. state, ARGV[1]) == 0 then
-		return redis.error_reply("task is not found in zset: " .. tostring(state))
+		return redis.error_reply("task is not found in zset: " .. tostring(ARGV[2] .. state))
 	end
 end
 local unique_key = redis.call("HGET", KEYS[1], "unique_key")
@@ -1214,10 +1469,12 @@ func (r *RDB) DeleteTask(qname, id string) error {
 	}
 	keys := []string{
 		base.TaskKey(qname, id),
+		base.AllGroups(qname),
 	}
 	argv := []interface{}{
 		id,
 		base.QueueKeyPrefix(qname),
+		base.GroupKeyPrefix(qname),
 	}
 	res, err := deleteTaskCmd.Run(context.Background(), r.client, keys, argv...).Result()
 	if err != nil {
@@ -1336,6 +1593,50 @@ func (r *RDB) deleteAll(key, qname string) (int64, error) {
 	return n, nil
 }
 
+// deleteAllAggregatingCmd deletes all tasks from the given group.
+//
+// Input:
+// KEYS[1] -> asynq:{<qname>}:g:<gname>
+// KEYS[2] -> asynq:{<qname>}:groups
+// -------
+// ARGV[1] -> task key prefix
+// ARGV[2] -> group name
+var deleteAllAggregatingCmd = redis.NewScript(`
+local ids = redis.call("ZRANGE", KEYS[1], 0, -1)
+for _, id in ipairs(ids) do
+	redis.call("DEL", ARGV[1] .. id)
+end
+redis.call("SREM", KEYS[2], ARGV[2])
+redis.call("DEL", KEYS[1])
+return table.getn(ids)
+`)
+
+// DeleteAllAggregatingTasks deletes all aggregating tasks from the given group
+// and returns the number of tasks deleted.
+func (r *RDB) DeleteAllAggregatingTasks(qname, gname string) (int64, error) {
+	var op errors.Op = "rdb.DeleteAllAggregatingTasks"
+	if err := r.checkQueueExists(qname); err != nil {
+		return 0, errors.E(op, errors.CanonicalCode(err), err)
+	}
+	keys := []string{
+		base.GroupKey(qname, gname),
+		base.AllGroups(qname),
+	}
+	argv := []interface{}{
+		base.TaskKeyPrefix(qname),
+		gname,
+	}
+	res, err := deleteAllAggregatingCmd.Run(context.Background(), r.client, keys, argv...).Result()
+	if err != nil {
+		return 0, errors.E(op, errors.Unknown, err)
+	}
+	n, ok := res.(int64)
+	if !ok {
+		return 0, errors.E(op, errors.Internal, "command error: unexpected return value %v", res)
+	}
+	return n, nil
+}
+
 // deleteAllPendingCmd deletes all pending tasks from the given queue.
 //
 // Input:

internal/rdb/rdb.go

@@ -12,6 +12,7 @@ import (
 	"time"
 
 	"github.com/go-redis/redis/v8"
+	"github.com/google/uuid"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/errors"
 	"github.com/hibiken/asynq/internal/timeutil"
@@ -348,7 +349,7 @@ func (r *RDB) Done(ctx context.Context, msg *base.TaskMessage) error {
 	argv := []interface{}{
 		msg.ID,
 		expireAt.Unix(),
-		math.MaxInt64,
+		int64(math.MaxInt64),
 	}
 	// Note: We cannot pass empty unique key when running this script in redis-cluster.
 	if len(msg.UniqueKey) > 0 {
@@ -458,7 +459,7 @@ func (r *RDB) MarkAsComplete(ctx context.Context, msg *base.TaskMessage) error {
 		statsExpireAt.Unix(),
 		now.Unix() + msg.Retention,
 		encoded,
-		math.MaxInt64,
+		int64(math.MaxInt64),
 	}
 	// Note: We cannot pass empty unique key when running this script in redis-cluster.
 	if len(msg.UniqueKey) > 0 {
@@ -497,6 +498,128 @@ func (r *RDB) Requeue(ctx context.Context, msg *base.TaskMessage) error {
 	return r.runScript(ctx, op, requeueCmd, keys, msg.ID)
 }
 
+// KEYS[1] -> asynq:{<qname>}:t:<task_id>
+// KEYS[2] -> asynq:{<qname>}:g:<group_key>
+// KEYS[3] -> asynq:{<qname>}:groups
+// -------
+// ARGV[1] -> task message data
+// ARGV[2] -> task ID
+// ARGV[3] -> current time in Unix time
+// ARGV[4] -> group key
+//
+// Output:
+// Returns 1 if successfully added
+// Returns 0 if task ID already exists
+var addToGroupCmd = redis.NewScript(`
+if redis.call("EXISTS", KEYS[1]) == 1 then
+	return 0
+end
+redis.call("HSET", KEYS[1],
+           "msg", ARGV[1],
+           "state", "aggregating",
+	       "group", ARGV[4])
+redis.call("ZADD", KEYS[2], ARGV[3], ARGV[2])
+redis.call("SADD", KEYS[3], ARGV[4])
+return 1
+`)
+
+func (r *RDB) AddToGroup(ctx context.Context, msg *base.TaskMessage, groupKey string) error {
+	var op errors.Op = "rdb.AddToGroup"
+	encoded, err := base.EncodeMessage(msg)
+	if err != nil {
+		return errors.E(op, errors.Unknown, fmt.Sprintf("cannot encode message: %v", err))
+	}
+	if err := r.client.SAdd(ctx, base.AllQueues, msg.Queue).Err(); err != nil {
+		return errors.E(op, errors.Unknown, &errors.RedisCommandError{Command: "sadd", Err: err})
+	}
+	keys := []string{
+		base.TaskKey(msg.Queue, msg.ID),
+		base.GroupKey(msg.Queue, groupKey),
+		base.AllGroups(msg.Queue),
+	}
+	argv := []interface{}{
+		encoded,
+		msg.ID,
+		r.clock.Now().Unix(),
+		groupKey,
+	}
+	n, err := r.runScriptWithErrorCode(ctx, op, addToGroupCmd, keys, argv...)
+	if err != nil {
+		return err
+	}
+	if n == 0 {
+		return errors.E(op, errors.AlreadyExists, errors.ErrTaskIdConflict)
+	}
+	return nil
+}
+
+// KEYS[1] -> asynq:{<qname>}:t:<task_id>
+// KEYS[2] -> asynq:{<qname>}:g:<group_key>
+// KEYS[3] -> asynq:{<qname>}:groups
+// KEYS[4] -> unique key
+// -------
+// ARGV[1] -> task message data
+// ARGV[2] -> task ID
+// ARGV[3] -> current time in Unix time
+// ARGV[4] -> group key
+// ARGV[5] -> uniqueness lock TTL
+//
+// Output:
+// Returns 1 if successfully added
+// Returns 0 if task ID already exists
+// Returns -1 if task unique key already exists
+var addToGroupUniqueCmd = redis.NewScript(`
+local ok = redis.call("SET", KEYS[4], ARGV[2], "NX", "EX", ARGV[5])
+if not ok then
+  return -1
+end
+if redis.call("EXISTS", KEYS[1]) == 1 then
+	return 0
+end
+redis.call("HSET", KEYS[1],
+           "msg", ARGV[1],
+           "state", "aggregating",
+	       "group", ARGV[4])
+redis.call("ZADD", KEYS[2], ARGV[3], ARGV[2])
+redis.call("SADD", KEYS[3], ARGV[4])
+return 1
+`)
+
+func (r *RDB) AddToGroupUnique(ctx context.Context, msg *base.TaskMessage, groupKey string, ttl time.Duration) error {
+	var op errors.Op = "rdb.AddToGroupUnique"
+	encoded, err := base.EncodeMessage(msg)
+	if err != nil {
+		return errors.E(op, errors.Unknown, fmt.Sprintf("cannot encode message: %v", err))
+	}
+	if err := r.client.SAdd(ctx, base.AllQueues, msg.Queue).Err(); err != nil {
+		return errors.E(op, errors.Unknown, &errors.RedisCommandError{Command: "sadd", Err: err})
+	}
+	keys := []string{
+		base.TaskKey(msg.Queue, msg.ID),
+		base.GroupKey(msg.Queue, groupKey),
+		base.AllGroups(msg.Queue),
+		base.UniqueKey(msg.Queue, msg.Type, msg.Payload),
+	}
+	argv := []interface{}{
+		encoded,
+		msg.ID,
+		r.clock.Now().Unix(),
+		groupKey,
+		int(ttl.Seconds()),
+	}
+	n, err := r.runScriptWithErrorCode(ctx, op, addToGroupUniqueCmd, keys, argv...)
+	if err != nil {
+		return err
+	}
+	if n == -1 {
+		return errors.E(op, errors.AlreadyExists, errors.ErrDuplicateTask)
+	}
+	if n == 0 {
+		return errors.E(op, errors.AlreadyExists, errors.ErrTaskIdConflict)
+	}
+	return nil
+}
+
 // KEYS[1] -> asynq:{<qname>}:t:<task_id>
 // KEYS[2] -> asynq:{<qname>}:scheduled
 // -------
@@ -688,7 +811,7 @@ func (r *RDB) Retry(ctx context.Context, msg *base.TaskMessage, processAt time.T
 		processAt.Unix(),
 		expireAt.Unix(),
 		isFailure,
-		math.MaxInt64,
+		int64(math.MaxInt64),
 	}
 	return r.runScript(ctx, op, retryCmd, keys, argv...)
 }
@@ -774,7 +897,7 @@ func (r *RDB) Archive(ctx context.Context, msg *base.TaskMessage, errMsg string)
 		cutoff.Unix(),
 		maxArchiveSize,
 		expireAt.Unix(),
-		math.MaxInt64,
+		int64(math.MaxInt64),
 	}
 	return r.runScript(ctx, op, archiveCmd, keys, argv...)
 }
@@ -796,24 +919,41 @@ func (r *RDB) ForwardIfReady(qnames ...string) error {
 // ARGV[1] -> current unix time in seconds
 // ARGV[2] -> task key prefix
 // ARGV[3] -> current unix time in nsec
+// ARGV[4] -> group key prefix
 // Note: Script moves tasks up to 100 at a time to keep the runtime of script short.
 var forwardCmd = redis.NewScript(`
 local ids = redis.call("ZRANGEBYSCORE", KEYS[1], "-inf", ARGV[1], "LIMIT", 0, 100)
 for _, id in ipairs(ids) do
-	redis.call("LPUSH", KEYS[2], id)
-	redis.call("ZREM", KEYS[1], id)
-	redis.call("HSET", ARGV[2] .. id,
-               "state", "pending",
-               "pending_since", ARGV[3])
+	local taskKey = ARGV[2] .. id
+	local group = redis.call("HGET", taskKey, "group")
+	if group and group ~= '' then
+	    redis.call("ZADD", ARGV[4] .. group, ARGV[1], id)
+		redis.call("ZREM", KEYS[1], id)
+		redis.call("HSET", taskKey,
+				   "state", "aggregating")
+	else
+		redis.call("LPUSH", KEYS[2], id)
+		redis.call("ZREM", KEYS[1], id)
+		redis.call("HSET", taskKey,
+				   "state", "pending",
+				   "pending_since", ARGV[3])
+	end
 end
 return table.getn(ids)`)
 
-// forward moves tasks with a score less than the current unix time
-// from the src zset to the dst list. It returns the number of tasks moved.
-func (r *RDB) forward(src, dst, taskKeyPrefix string) (int, error) {
+// forward moves tasks with a score less than the current unix time from the delayed (i.e. scheduled | retry) zset
+// to the pending list or group set.
+// It returns the number of tasks moved.
+func (r *RDB) forward(delayedKey, pendingKey, taskKeyPrefix, groupKeyPrefix string) (int, error) {
 	now := r.clock.Now()
-	res, err := forwardCmd.Run(context.Background(), r.client,
-		[]string{src, dst}, now.Unix(), taskKeyPrefix, now.UnixNano()).Result()
+	keys := []string{delayedKey, pendingKey}
+	argv := []interface{}{
+		now.Unix(),
+		taskKeyPrefix,
+		now.UnixNano(),
+		groupKeyPrefix,
+	}
+	res, err := forwardCmd.Run(context.Background(), r.client, keys, argv...).Result()
 	if err != nil {
 		return 0, errors.E(errors.Internal, fmt.Sprintf("redis eval error: %v", err))
 	}
@@ -825,15 +965,16 @@ func (r *RDB) forward(src, dst, taskKeyPrefix string) (int, error) {
 }
 
 // forwardAll checks for tasks in scheduled/retry state that are ready to be run, and updates
-// their state to "pending".
+// their state to "pending" or "aggregating".
 func (r *RDB) forwardAll(qname string) (err error) {
-	sources := []string{base.ScheduledKey(qname), base.RetryKey(qname)}
-	dst := base.PendingKey(qname)
+	delayedKeys := []string{base.ScheduledKey(qname), base.RetryKey(qname)}
+	pendingKey := base.PendingKey(qname)
 	taskKeyPrefix := base.TaskKeyPrefix(qname)
-	for _, src := range sources {
+	groupKeyPrefix := base.GroupKeyPrefix(qname)
+	for _, delayedKey := range delayedKeys {
 		n := 1
 		for n != 0 {
-			n, err = r.forward(src, dst, taskKeyPrefix)
+			n, err = r.forward(delayedKey, pendingKey, taskKeyPrefix, groupKeyPrefix)
 			if err != nil {
 				return err
 			}
@@ -842,6 +983,248 @@ func (r *RDB) forwardAll(qname string) (err error) {
 	return nil
 }
 
+// ListGroups returns a list of all known groups in the given queue.
+func (r *RDB) ListGroups(qname string) ([]string, error) {
+	var op errors.Op = "RDB.ListGroups"
+	groups, err := r.client.SMembers(context.Background(), base.AllGroups(qname)).Result()
+	if err != nil {
+		return nil, errors.E(op, errors.Unknown, &errors.RedisCommandError{Command: "smembers", Err: err})
+	}
+	return groups, nil
+}
+
+// aggregationCheckCmd checks the given group for whether to create an aggregation set.
+// An aggregation set is created if one of the aggregation criteria is met:
+// 1) group has reached or exceeded its max size
+// 2) group's oldest task has reached or exceeded its max delay
+// 3) group's latest task has reached or exceeded its grace period
+// if aggreation criteria is met, the command moves those tasks from the group
+// and put them in an aggregation set. Additionally, if the creation of aggregation set
+// empties the group, it will clear the group name from the all groups set.
+//
+// KEYS[1] -> asynq:{<qname>}:g:<gname>
+// KEYS[2] -> asynq:{<qname>}:g:<gname>:<aggregation_set_id>
+// KEYS[3] -> asynq:{<qname>}:aggregation_sets
+// KEYS[4] -> asynq:{<qname>}:groups
+// -------
+// ARGV[1] -> max group size
+// ARGV[2] -> max group delay in unix time
+// ARGV[3] -> start time of the grace period
+// ARGV[4] -> aggregation set expire time
+// ARGV[5] -> current time in unix time
+// ARGV[6] -> group name
+//
+// Output:
+// Returns 0 if no aggregation set was created
+// Returns 1 if an aggregation set was created
+//
+// Time Complexity:
+// O(log(N) + M) with N being the number tasks in the group zset
+// and M being the max size.
+var aggregationCheckCmd = redis.NewScript(`
+local size = redis.call("ZCARD", KEYS[1])
+if size == 0 then
+	return 0
+end
+local maxSize = tonumber(ARGV[1])
+if maxSize ~= 0 and size >= maxSize then
+	local res = redis.call("ZRANGE", KEYS[1], 0, maxSize-1, "WITHSCORES")
+	for i=1, table.getn(res)-1, 2 do
+		redis.call("ZADD", KEYS[2], tonumber(res[i+1]), res[i])
+	end
+	redis.call("ZREMRANGEBYRANK", KEYS[1], 0, maxSize-1)
+	redis.call("ZADD", KEYS[3], ARGV[4], KEYS[2])
+	if size == maxSize then
+		redis.call("SREM", KEYS[4], ARGV[6])
+	end
+	return 1
+end
+local maxDelay = tonumber(ARGV[2])
+local currentTime = tonumber(ARGV[5])
+if maxDelay ~= 0 then
+	local oldestEntry = redis.call("ZRANGE", KEYS[1], 0, 0, "WITHSCORES")
+	local oldestEntryScore = tonumber(oldestEntry[2])
+	local maxDelayTime = currentTime - maxDelay
+	if oldestEntryScore <= maxDelayTime then
+		local res = redis.call("ZRANGE", KEYS[1], 0, maxSize-1, "WITHSCORES")
+		for i=1, table.getn(res)-1, 2 do
+			redis.call("ZADD", KEYS[2], tonumber(res[i+1]), res[i])
+		end
+		redis.call("ZREMRANGEBYRANK", KEYS[1], 0, maxSize-1)
+		redis.call("ZADD", KEYS[3], ARGV[4], KEYS[2])
+		if size <= maxSize or maxSize == 0 then
+			redis.call("SREM", KEYS[4], ARGV[6])
+		end
+		return 1
+	end
+end
+local latestEntry = redis.call("ZREVRANGE", KEYS[1], 0, 0, "WITHSCORES")
+local latestEntryScore = tonumber(latestEntry[2])
+local gracePeriodStartTime = currentTime - tonumber(ARGV[3])
+if latestEntryScore <= gracePeriodStartTime then
+	local res = redis.call("ZRANGE", KEYS[1], 0, maxSize-1, "WITHSCORES")
+	for i=1, table.getn(res)-1, 2 do
+		redis.call("ZADD", KEYS[2], tonumber(res[i+1]), res[i])
+	end
+	redis.call("ZREMRANGEBYRANK", KEYS[1], 0, maxSize-1)
+	redis.call("ZADD", KEYS[3], ARGV[4], KEYS[2])
+	if size <= maxSize or maxSize == 0 then
+		redis.call("SREM", KEYS[4], ARGV[6])
+	end
+	return 1
+end
+return 0
+`)
+
+// Task aggregation should finish within this timeout.
+// Otherwise an aggregation set should be reclaimed by the recoverer.
+const aggregationTimeout = 2 * time.Minute
+
+// AggregationCheck checks the group identified by the given queue and group name to see if the tasks in the
+// group are ready to be aggregated. If so, it moves the tasks to be aggregated to a aggregation set and returns
+// the set ID. If not, it returns an empty string for the set ID.
+// The time for gracePeriod and maxDelay is computed relative to the time t.
+//
+// Note: It assumes that this function is called at frequency less than or equal to the gracePeriod. In other words,
+// the function only checks the most recently added task aganist the given gracePeriod.
+func (r *RDB) AggregationCheck(qname, gname string, t time.Time, gracePeriod, maxDelay time.Duration, maxSize int) (string, error) {
+	var op errors.Op = "RDB.AggregationCheck"
+	aggregationSetID := uuid.NewString()
+	expireTime := r.clock.Now().Add(aggregationTimeout)
+	keys := []string{
+		base.GroupKey(qname, gname),
+		base.AggregationSetKey(qname, gname, aggregationSetID),
+		base.AllAggregationSets(qname),
+		base.AllGroups(qname),
+	}
+	argv := []interface{}{
+		maxSize,
+		int64(maxDelay.Seconds()),
+		int64(gracePeriod.Seconds()),
+		expireTime.Unix(),
+		t.Unix(),
+		gname,
+	}
+	n, err := r.runScriptWithErrorCode(context.Background(), op, aggregationCheckCmd, keys, argv...)
+	if err != nil {
+		return "", err
+	}
+	switch n {
+	case 0:
+		return "", nil
+	case 1:
+		return aggregationSetID, nil
+	default:
+		return "", errors.E(op, errors.Internal, fmt.Sprintf("unexpected return value from lua script: %d", n))
+	}
+}
+
+// KEYS[1] -> asynq:{<qname>}:g:<gname>:<aggregation_set_id>
+// ------
+// ARGV[1] -> task key prefix
+//
+// Output:
+// Array of encoded task messages
+//
+// Time Complexity:
+// O(N) with N being the number of tasks in the aggregation set.
+var readAggregationSetCmd = redis.NewScript(`
+local msgs = {}
+local ids = redis.call("ZRANGE", KEYS[1], 0, -1)
+for _, id in ipairs(ids) do
+	local key = ARGV[1] .. id
+	table.insert(msgs, redis.call("HGET", key, "msg"))
+end
+return msgs
+`)
+
+// ReadAggregationSet retrieves members of an aggregation set and returns a list of tasks in the set and
+// the deadline for aggregating those tasks.
+func (r *RDB) ReadAggregationSet(qname, gname, setID string) ([]*base.TaskMessage, time.Time, error) {
+	var op errors.Op = "RDB.ReadAggregationSet"
+	ctx := context.Background()
+	aggSetKey := base.AggregationSetKey(qname, gname, setID)
+	res, err := readAggregationSetCmd.Run(ctx, r.client,
+		[]string{aggSetKey}, base.TaskKeyPrefix(qname)).Result()
+	if err != nil {
+		return nil, time.Time{}, errors.E(op, errors.Unknown, fmt.Sprintf("redis eval error: %v", err))
+	}
+	data, err := cast.ToStringSliceE(res)
+	if err != nil {
+		return nil, time.Time{}, errors.E(op, errors.Internal, fmt.Sprintf("cast error: Lua script returned unexpected value: %v", res))
+	}
+	var msgs []*base.TaskMessage
+	for _, s := range data {
+		msg, err := base.DecodeMessage([]byte(s))
+		if err != nil {
+			return nil, time.Time{}, errors.E(op, errors.Internal, fmt.Sprintf("cannot decode message: %v", err))
+		}
+		msgs = append(msgs, msg)
+	}
+	deadlineUnix, err := r.client.ZScore(ctx, base.AllAggregationSets(qname), aggSetKey).Result()
+	if err != nil {
+		return nil, time.Time{}, errors.E(op, errors.Unknown, &errors.RedisCommandError{Command: "zscore", Err: err})
+	}
+	return msgs, time.Unix(int64(deadlineUnix), 0), nil
+}
+
+// KEYS[1] -> asynq:{<qname>}:g:<gname>:<aggregation_set_id>
+// KEYS[2] -> asynq:{<qname>}:aggregation_sets
+// -------
+// ARGV[1] -> task key prefix
+//
+// Output:
+// Redis status reply
+//
+// Time Complexity:
+// max(O(N), O(log(M))) with N being the number of tasks in the aggregation set
+// and M being the number of elements in the all-aggregation-sets list.
+var deleteAggregationSetCmd = redis.NewScript(`
+local ids = redis.call("ZRANGE", KEYS[1], 0, -1)
+for _, id in ipairs(ids)  do
+	redis.call("DEL", ARGV[1] .. id)
+end
+redis.call("DEL", KEYS[1])
+redis.call("ZREM", KEYS[2], KEYS[1])
+return redis.status_reply("OK")
+`)
+
+// DeleteAggregationSet deletes the aggregation set and its members identified by the parameters.
+func (r *RDB) DeleteAggregationSet(ctx context.Context, qname, gname, setID string) error {
+	var op errors.Op = "RDB.DeleteAggregationSet"
+	keys := []string{
+		base.AggregationSetKey(qname, gname, setID),
+		base.AllAggregationSets(qname),
+	}
+	return r.runScript(ctx, op, deleteAggregationSetCmd, keys, base.TaskKeyPrefix(qname))
+}
+
+// KEYS[1] -> asynq:{<qname>}:aggregation_sets
+// -------
+// ARGV[1] -> current time in unix time
+var reclaimStateAggregationSetsCmd = redis.NewScript(`
+local staleSetKeys = redis.call("ZRANGEBYSCORE", KEYS[1], "-inf", ARGV[1])
+for _, key in ipairs(staleSetKeys) do
+	local idx = string.find(key, ":[^:]*$")
+	local groupKey = string.sub(key, 1, idx-1)
+	local res = redis.call("ZRANGE", key, 0, -1, "WITHSCORES")
+	for i=1, table.getn(res)-1, 2 do
+		redis.call("ZADD", groupKey, tonumber(res[i+1]), res[i])
+	end
+	redis.call("DEL", key)
+end
+redis.call("ZREMRANGEBYSCORE", KEYS[1], "-inf", ARGV[1])
+return redis.status_reply("OK")
+`)
+
+// ReclaimStateAggregationSets checks for any stale aggregation sets in the given queue, and
+// reclaim tasks in the stale aggregation set by putting them back in the group.
+func (r *RDB) ReclaimStaleAggregationSets(qname string) error {
+	var op errors.Op = "RDB.ReclaimStaleAggregationSets"
+	return r.runScript(context.Background(), op, reclaimStateAggregationSetsCmd,
+		[]string{base.AllAggregationSets(qname)}, r.clock.Now().Unix())
+}
+
 // KEYS[1] -> asynq:{<qname>}:completed
 // ARGV[1] -> current time in unix time
 // ARGV[2] -> task key prefix

internal/testbroker/testbroker.go

@@ -15,7 +15,7 @@ import (
 	"github.com/hibiken/asynq/internal/base"
 )
 
-var errRedisDown = errors.New("asynqtest: redis is down")
+var errRedisDown = errors.New("testutil: redis is down")
 
 // TestBroker is a broker implementation which enables
 // to simulate Redis failure in tests.
@@ -234,3 +234,66 @@ func (tb *TestBroker) Close() error {
 	}
 	return tb.real.Close()
 }
+
+func (tb *TestBroker) AddToGroup(ctx context.Context, msg *base.TaskMessage, gname string) error {
+	tb.mu.Lock()
+	defer tb.mu.Unlock()
+	if tb.sleeping {
+		return errRedisDown
+	}
+	return tb.real.AddToGroup(ctx, msg, gname)
+}
+
+func (tb *TestBroker) AddToGroupUnique(ctx context.Context, msg *base.TaskMessage, gname string, ttl time.Duration) error {
+	tb.mu.Lock()
+	defer tb.mu.Unlock()
+	if tb.sleeping {
+		return errRedisDown
+	}
+	return tb.real.AddToGroupUnique(ctx, msg, gname, ttl)
+}
+
+func (tb *TestBroker) ListGroups(qname string) ([]string, error) {
+	tb.mu.Lock()
+	defer tb.mu.Unlock()
+	if tb.sleeping {
+		return nil, errRedisDown
+	}
+	return tb.real.ListGroups(qname)
+}
+
+func (tb *TestBroker) AggregationCheck(qname, gname string, t time.Time, gracePeriod, maxDelay time.Duration, maxSize int) (aggregationSetID string, err error) {
+	tb.mu.Lock()
+	defer tb.mu.Unlock()
+	if tb.sleeping {
+		return "", errRedisDown
+	}
+	return tb.real.AggregationCheck(qname, gname, t, gracePeriod, maxDelay, maxSize)
+}
+
+func (tb *TestBroker) ReadAggregationSet(qname, gname, aggregationSetID string) ([]*base.TaskMessage, time.Time, error) {
+	tb.mu.Lock()
+	defer tb.mu.Unlock()
+	if tb.sleeping {
+		return nil, time.Time{}, errRedisDown
+	}
+	return tb.real.ReadAggregationSet(qname, gname, aggregationSetID)
+}
+
+func (tb *TestBroker) DeleteAggregationSet(ctx context.Context, qname, gname, aggregationSetID string) error {
+	tb.mu.Lock()
+	defer tb.mu.Unlock()
+	if tb.sleeping {
+		return errRedisDown
+	}
+	return tb.real.DeleteAggregationSet(ctx, qname, gname, aggregationSetID)
+}
+
+func (tb *TestBroker) ReclaimStaleAggregationSets(qname string) error {
+	tb.mu.Lock()
+	defer tb.mu.Unlock()
+	if tb.sleeping {
+		return errRedisDown
+	}
+	return tb.real.ReclaimStaleAggregationSets(qname)
+}

internal/testutil/builder.go

@@ -0,0 +1,84 @@
+// Copyright 2022 Kentaro Hibino. All rights reserved.
+// Use of this source code is governed by a MIT license
+// that can be found in the LICENSE file.
+
+package testutil
+
+import (
+	"time"
+
+	"github.com/google/uuid"
+	"github.com/hibiken/asynq/internal/base"
+)
+
+func makeDefaultTaskMessage() *base.TaskMessage {
+	return &base.TaskMessage{
+		ID:       uuid.NewString(),
+		Type:     "default_task",
+		Queue:    "default",
+		Retry:    25,
+		Timeout:  1800, // default timeout of 30 mins
+		Deadline: 0,    // no deadline
+	}
+}
+
+type TaskMessageBuilder struct {
+	msg *base.TaskMessage
+}
+
+func NewTaskMessageBuilder() *TaskMessageBuilder {
+	return &TaskMessageBuilder{}
+}
+
+func (b *TaskMessageBuilder) lazyInit() {
+	if b.msg == nil {
+		b.msg = makeDefaultTaskMessage()
+	}
+}
+
+func (b *TaskMessageBuilder) Build() *base.TaskMessage {
+	b.lazyInit()
+	return b.msg
+}
+
+func (b *TaskMessageBuilder) SetType(typename string) *TaskMessageBuilder {
+	b.lazyInit()
+	b.msg.Type = typename
+	return b
+}
+
+func (b *TaskMessageBuilder) SetPayload(payload []byte) *TaskMessageBuilder {
+	b.lazyInit()
+	b.msg.Payload = payload
+	return b
+}
+
+func (b *TaskMessageBuilder) SetQueue(qname string) *TaskMessageBuilder {
+	b.lazyInit()
+	b.msg.Queue = qname
+	return b
+}
+
+func (b *TaskMessageBuilder) SetRetry(n int) *TaskMessageBuilder {
+	b.lazyInit()
+	b.msg.Retry = n
+	return b
+}
+
+func (b *TaskMessageBuilder) SetTimeout(timeout time.Duration) *TaskMessageBuilder {
+	b.lazyInit()
+	b.msg.Timeout = int64(timeout.Seconds())
+	return b
+}
+
+func (b *TaskMessageBuilder) SetDeadline(deadline time.Time) *TaskMessageBuilder {
+	b.lazyInit()
+	b.msg.Deadline = deadline.Unix()
+	return b
+}
+
+func (b *TaskMessageBuilder) SetGroup(gname string) *TaskMessageBuilder {
+	b.lazyInit()
+	b.msg.GroupKey = gname
+	return b
+}

internal/testutil/builder_test.go

@@ -0,0 +1,94 @@
+// Copyright 2022 Kentaro Hibino. All rights reserved.
+// Use of this source code is governed by a MIT license
+// that can be found in the LICENSE file.
+
+package testutil
+
+import (
+	"testing"
+	"time"
+
+	"github.com/google/go-cmp/cmp"
+	"github.com/google/go-cmp/cmp/cmpopts"
+	"github.com/hibiken/asynq/internal/base"
+)
+
+func TestTaskMessageBuilder(t *testing.T) {
+	tests := []struct {
+		desc string
+		ops  func(b *TaskMessageBuilder) // operations to perform on the builder
+		want *base.TaskMessage
+	}{
+		{
+			desc: "zero value and build",
+			ops:  nil,
+			want: &base.TaskMessage{
+				Type:     "default_task",
+				Queue:    "default",
+				Payload:  nil,
+				Retry:    25,
+				Timeout:  1800, // 30m
+				Deadline: 0,
+			},
+		},
+		{
+			desc: "with type, payload, and queue",
+			ops: func(b *TaskMessageBuilder) {
+				b.SetType("foo").SetPayload([]byte("hello")).SetQueue("myqueue")
+			},
+			want: &base.TaskMessage{
+				Type:     "foo",
+				Queue:    "myqueue",
+				Payload:  []byte("hello"),
+				Retry:    25,
+				Timeout:  1800, // 30m
+				Deadline: 0,
+			},
+		},
+		{
+			desc: "with retry, timeout, and deadline",
+			ops: func(b *TaskMessageBuilder) {
+				b.SetRetry(1).
+					SetTimeout(20 * time.Second).
+					SetDeadline(time.Date(2017, 3, 6, 0, 0, 0, 0, time.UTC))
+			},
+			want: &base.TaskMessage{
+				Type:     "default_task",
+				Queue:    "default",
+				Payload:  nil,
+				Retry:    1,
+				Timeout:  20,
+				Deadline: time.Date(2017, 3, 6, 0, 0, 0, 0, time.UTC).Unix(),
+			},
+		},
+		{
+			desc: "with group",
+			ops: func(b *TaskMessageBuilder) {
+				b.SetGroup("mygroup")
+			},
+			want: &base.TaskMessage{
+				Type:     "default_task",
+				Queue:    "default",
+				Payload:  nil,
+				Retry:    25,
+				Timeout:  1800,
+				Deadline: 0,
+				GroupKey: "mygroup",
+			},
+		},
+	}
+	cmpOpts := []cmp.Option{cmpopts.IgnoreFields(base.TaskMessage{}, "ID")}
+
+	for _, tc := range tests {
+		var b TaskMessageBuilder
+		if tc.ops != nil {
+			tc.ops(&b)
+		}
+
+		got := b.Build()
+		if diff := cmp.Diff(tc.want, got, cmpOpts...); diff != "" {
+			t.Errorf("%s: TaskMessageBuilder.Build() = %+v, want %+v;\n(-want,+got)\n%s",
+				tc.desc, got, tc.want, diff)
+		}
+	}
+}

internal/testutil/testutil.go

@@ -2,8 +2,8 @@
 // Use of this source code is governed by a MIT license
 // that can be found in the LICENSE file.
 
-// Package asynqtest defines test helpers for asynq and its internal packages.
-package asynqtest
+// Package testutil defines test helpers for asynq and its internal packages.
+package testutil
 
 import (
 	"context"
@@ -93,6 +93,20 @@ var SortStringSliceOpt = cmp.Transformer("SortStringSlice", func(in []string) []
 	return out
 })
 
+var SortRedisZSetEntryOpt = cmp.Transformer("SortZSetEntries", func(in []redis.Z) []redis.Z {
+	out := append([]redis.Z(nil), in...) // Copy input to avoid mutating it
+	sort.Slice(out, func(i, j int) bool {
+		// TODO: If member is a comparable type (int, string, etc) compare by the member
+		// Use generic comparable type here once update to go1.18
+		if _, ok := out[i].Member.(string); ok {
+			// If member is a string, compare the member
+			return out[i].Member.(string) < out[j].Member.(string)
+		}
+		return out[i].Score < out[j].Score
+	})
+	return out
+})
+
 // IgnoreIDOpt is an cmp.Option to ignore ID field in task messages when comparing.
 var IgnoreIDOpt = cmpopts.IgnoreFields(base.TaskMessage{}, "ID")
 
@@ -245,6 +259,21 @@ func SeedCompletedQueue(tb testing.TB, r redis.UniversalClient, entries []base.Z
 	seedRedisZSet(tb, r, base.CompletedKey(qname), entries, base.TaskStateCompleted)
 }
 
+// SeedGroup initializes the group with the given entries.
+func SeedGroup(tb testing.TB, r redis.UniversalClient, entries []base.Z, qname, gname string) {
+	tb.Helper()
+	ctx := context.Background()
+	r.SAdd(ctx, base.AllQueues, qname)
+	r.SAdd(ctx, base.AllGroups(qname), gname)
+	seedRedisZSet(tb, r, base.GroupKey(qname, gname), entries, base.TaskStateAggregating)
+}
+
+func SeedAggregationSet(tb testing.TB, r redis.UniversalClient, entries []base.Z, qname, gname, setID string) {
+	tb.Helper()
+	r.SAdd(context.Background(), base.AllQueues, qname)
+	seedRedisZSet(tb, r, base.AggregationSetKey(qname, gname, setID), entries, base.TaskStateAggregating)
+}
+
 // SeedAllPendingQueues initializes all of the specified queues with the given messages.
 //
 // pending maps a queue name to a list of messages.
@@ -303,6 +332,18 @@ func SeedAllCompletedQueues(tb testing.TB, r redis.UniversalClient, completed ma
 	}
 }
 
+// SeedAllGroups initializes all groups in all queues.
+// The map maps queue names to group names which maps to a list of task messages and the time it was
+// added to the group.
+func SeedAllGroups(tb testing.TB, r redis.UniversalClient, groups map[string]map[string][]base.Z) {
+	tb.Helper()
+	for qname, g := range groups {
+		for gname, entries := range g {
+			SeedGroup(tb, r, entries, qname, gname)
+		}
+	}
+}
+
 func seedRedisList(tb testing.TB, c redis.UniversalClient, key string,
 	msgs []*base.TaskMessage, state base.TaskState) {
 	tb.Helper()
@@ -311,13 +352,14 @@ func seedRedisList(tb testing.TB, c redis.UniversalClient, key string,
 		if err := c.LPush(context.Background(), key, msg.ID).Err(); err != nil {
 			tb.Fatal(err)
 		}
-		key := base.TaskKey(msg.Queue, msg.ID)
+		taskKey := base.TaskKey(msg.Queue, msg.ID)
 		data := map[string]interface{}{
 			"msg":        encoded,
 			"state":      state.String(),
 			"unique_key": msg.UniqueKey,
+			"group":      msg.GroupKey,
 		}
-		if err := c.HSet(context.Background(), key, data).Err(); err != nil {
+		if err := c.HSet(context.Background(), taskKey, data).Err(); err != nil {
 			tb.Fatal(err)
 		}
 		if len(msg.UniqueKey) > 0 {
@@ -339,13 +381,14 @@ func seedRedisZSet(tb testing.TB, c redis.UniversalClient, key string,
 		if err := c.ZAdd(context.Background(), key, z).Err(); err != nil {
 			tb.Fatal(err)
 		}
-		key := base.TaskKey(msg.Queue, msg.ID)
+		taskKey := base.TaskKey(msg.Queue, msg.ID)
 		data := map[string]interface{}{
 			"msg":        encoded,
 			"state":      state.String(),
 			"unique_key": msg.UniqueKey,
+			"group":      msg.GroupKey,
 		}
-		if err := c.HSet(context.Background(), key, data).Err(); err != nil {
+		if err := c.HSet(context.Background(), taskKey, data).Err(); err != nil {
 			tb.Fatal(err)
 		}
 		if len(msg.UniqueKey) > 0 {
@@ -434,6 +477,14 @@ func GetCompletedEntries(tb testing.TB, r redis.UniversalClient, qname string) [
 	return getMessagesFromZSetWithScores(tb, r, qname, base.CompletedKey, base.TaskStateCompleted)
 }
 
+// GetGroupEntries returns all scheduled messages and its score in the given queue.
+// It also asserts the state field of the task.
+func GetGroupEntries(tb testing.TB, r redis.UniversalClient, qname, groupKey string) []base.Z {
+	tb.Helper()
+	return getMessagesFromZSetWithScores(tb, r, qname,
+		func(qname string) string { return base.GroupKey(qname, groupKey) }, base.TaskStateAggregating)
+}
+
 // Retrieves all messages stored under `keyFn(qname)` key in redis list.
 func getMessagesFromList(tb testing.TB, r redis.UniversalClient, qname string,
 	keyFn func(qname string) string, state base.TaskState) []*base.TaskMessage {
@@ -485,3 +536,107 @@ func getMessagesFromZSetWithScores(tb testing.TB, r redis.UniversalClient,
 	}
 	return res
 }
+
+// TaskSeedData holds the data required to seed tasks under the task key in test.
+type TaskSeedData struct {
+	Msg          *base.TaskMessage
+	State        base.TaskState
+	PendingSince time.Time
+}
+
+func SeedTasks(tb testing.TB, r redis.UniversalClient, taskData []*TaskSeedData) {
+	for _, data := range taskData {
+		msg := data.Msg
+		ctx := context.Background()
+		key := base.TaskKey(msg.Queue, msg.ID)
+		v := map[string]interface{}{
+			"msg":        MustMarshal(tb, msg),
+			"state":      data.State.String(),
+			"unique_key": msg.UniqueKey,
+			"group":      msg.GroupKey,
+		}
+		if !data.PendingSince.IsZero() {
+			v["pending_since"] = data.PendingSince.Unix()
+		}
+		if err := r.HSet(ctx, key, v).Err(); err != nil {
+			tb.Fatalf("Failed to write task data in redis: %v", err)
+		}
+		if len(msg.UniqueKey) > 0 {
+			err := r.SetNX(ctx, msg.UniqueKey, msg.ID, 1*time.Minute).Err()
+			if err != nil {
+				tb.Fatalf("Failed to set unique lock in redis: %v", err)
+			}
+		}
+	}
+}
+
+func SeedRedisZSets(tb testing.TB, r redis.UniversalClient, zsets map[string][]*redis.Z) {
+	for key, zs := range zsets {
+		// FIXME: How come we can't simply do ZAdd(ctx, key, zs...) here?
+		for _, z := range zs {
+			if err := r.ZAdd(context.Background(), key, z).Err(); err != nil {
+				tb.Fatalf("Failed to seed zset (key=%q): %v", key, err)
+			}
+		}
+	}
+}
+
+func SeedRedisSets(tb testing.TB, r redis.UniversalClient, sets map[string][]string) {
+	for key, set := range sets {
+		SeedRedisSet(tb, r, key, set)
+	}
+}
+
+func SeedRedisSet(tb testing.TB, r redis.UniversalClient, key string, members []string) {
+	for _, mem := range members {
+		if err := r.SAdd(context.Background(), key, mem).Err(); err != nil {
+			tb.Fatalf("Failed to seed set (key=%q): %v", key, err)
+		}
+	}
+}
+
+func SeedRedisLists(tb testing.TB, r redis.UniversalClient, lists map[string][]string) {
+	for key, vals := range lists {
+		for _, v := range vals {
+			if err := r.LPush(context.Background(), key, v).Err(); err != nil {
+				tb.Fatalf("Failed to seed list (key=%q): %v", key, err)
+			}
+		}
+	}
+}
+
+func AssertRedisLists(t *testing.T, r redis.UniversalClient, wantLists map[string][]string) {
+	for key, want := range wantLists {
+		got, err := r.LRange(context.Background(), key, 0, -1).Result()
+		if err != nil {
+			t.Fatalf("Failed to read list (key=%q): %v", key, err)
+		}
+		if diff := cmp.Diff(want, got, SortStringSliceOpt); diff != "" {
+			t.Errorf("mismatch found in list (key=%q): (-want,+got)\n%s", key, diff)
+		}
+	}
+}
+
+func AssertRedisSets(t *testing.T, r redis.UniversalClient, wantSets map[string][]string) {
+	for key, want := range wantSets {
+		got, err := r.SMembers(context.Background(), key).Result()
+		if err != nil {
+			t.Fatalf("Failed to read set (key=%q): %v", key, err)
+		}
+		if diff := cmp.Diff(want, got, SortStringSliceOpt); diff != "" {
+			t.Errorf("mismatch found in set (key=%q): (-want,+got)\n%s", key, diff)
+		}
+	}
+}
+
+func AssertRedisZSets(t *testing.T, r redis.UniversalClient, wantZSets map[string][]redis.Z) {
+	for key, want := range wantZSets {
+		got, err := r.ZRangeWithScores(context.Background(), key, 0, -1).Result()
+		if err != nil {
+			t.Fatalf("Failed to read zset (key=%q): %v", key, err)
+		}
+		if diff := cmp.Diff(want, got, SortRedisZSetEntryOpt); diff != "" {
+			t.Errorf("mismatch found in zset (key=%q): (-want,+got)\n%s", key, diff)
+		}
+	}
+}

janitor_test.go

@@ -10,9 +10,9 @@ import (
 	"time"
 
 	"github.com/google/go-cmp/cmp"
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/rdb"
+	h "github.com/hibiken/asynq/internal/testutil"
 )
 
 func newCompletedTask(qname, tasktype string, payload []byte, completedAt time.Time) *base.TaskMessage {

processor_test.go

@@ -15,11 +15,11 @@ import (
 
 	"github.com/google/go-cmp/cmp"
 	"github.com/google/go-cmp/cmp/cmpopts"
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/errors"
 	"github.com/hibiken/asynq/internal/log"
 	"github.com/hibiken/asynq/internal/rdb"
+	h "github.com/hibiken/asynq/internal/testutil"
 	"github.com/hibiken/asynq/internal/timeutil"
 )
 

recoverer.go

@@ -82,11 +82,16 @@ func (r *recoverer) start(wg *sync.WaitGroup) {
 var ErrLeaseExpired = errors.New("asynq: task lease expired")
 
 func (r *recoverer) recover() {
+	r.recoverLeaseExpiredTasks()
+	r.recoverStaleAggregationSets()
+}
+
+func (r *recoverer) recoverLeaseExpiredTasks() {
 	// Get all tasks which have expired 30 seconds ago or earlier to accomodate certain amount of clock skew.
 	cutoff := time.Now().Add(-30 * time.Second)
 	msgs, err := r.broker.ListLeaseExpired(cutoff, r.queues...)
 	if err != nil {
-		r.logger.Warn("recoverer: could not list lease expired tasks")
+		r.logger.Warnf("recoverer: could not list lease expired tasks: %v", err)
 		return
 	}
 	for _, msg := range msgs {
@@ -98,6 +103,14 @@ func (r *recoverer) recover() {
 	}
 }
 
+func (r *recoverer) recoverStaleAggregationSets() {
+	for _, qname := range r.queues {
+		if err := r.broker.ReclaimStaleAggregationSets(qname); err != nil {
+			r.logger.Warnf("recoverer: could not reclaim stale aggregation sets in queue %q: %v", qname, err)
+		}
+	}
+}
+
 func (r *recoverer) retry(msg *base.TaskMessage, err error) {
 	delay := r.retryDelayFunc(msg.Retried, err, NewTask(msg.Type, msg.Payload))
 	retryAt := time.Now().Add(delay)

recoverer_test.go

@@ -10,9 +10,9 @@ import (
 	"time"
 
 	"github.com/google/go-cmp/cmp"
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/rdb"
+	h "github.com/hibiken/asynq/internal/testutil"
 )
 
 func TestRecoverer(t *testing.T) {

scheduler_test.go

@@ -10,8 +10,8 @@ import (
 	"time"
 
 	"github.com/google/go-cmp/cmp"
-	"github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
+	"github.com/hibiken/asynq/internal/testutil"
 )
 
 func TestSchedulerRegister(t *testing.T) {
@@ -69,8 +69,8 @@ func TestSchedulerRegister(t *testing.T) {
 		time.Sleep(tc.wait)
 		scheduler.Shutdown()
 
-		got := asynqtest.GetPendingMessages(t, r, tc.queue)
-		if diff := cmp.Diff(tc.want, got, asynqtest.IgnoreIDOpt); diff != "" {
+		got := testutil.GetPendingMessages(t, r, tc.queue)
+		if diff := cmp.Diff(tc.want, got, testutil.IgnoreIDOpt); diff != "" {
 			t.Errorf("mismatch found in queue %q: (-want,+got)\n%s", tc.queue, diff)
 		}
 	}
@@ -148,7 +148,7 @@ func TestSchedulerUnregister(t *testing.T) {
 		time.Sleep(tc.wait)
 		scheduler.Shutdown()
 
-		got := asynqtest.GetPendingMessages(t, r, tc.queue)
+		got := testutil.GetPendingMessages(t, r, tc.queue)
 		if len(got) != 0 {
 			t.Errorf("%d tasks were enqueued, want zero", len(got))
 		}

server.go

@@ -50,6 +50,7 @@ type Server struct {
 	recoverer     *recoverer
 	healthchecker *healthchecker
 	janitor       *janitor
+	aggregator    *aggregator
 }
 
 type serverState struct {
@@ -193,6 +194,52 @@ type Config struct {
 	//
 	// If unset or zero, the interval is set to 5 seconds.
 	DelayedTaskCheckInterval time.Duration
+
+	// GroupGracePeriod specifies the amount of time the server will wait for an incoming task before aggregating
+	// the tasks in a group. If an incoming task is received within this period, the server will wait for another
+	// period of the same length, up to GroupMaxDelay if specified.
+	//
+	// If unset or zero, the grace period is set to 1 minute.
+	// Minimum duration for GroupGracePeriod is 1 second. If value specified is less than a second, the call to
+	// NewServer will panic.
+	GroupGracePeriod time.Duration
+
+	// GroupMaxDelay specifies the maximum amount of time the server will wait for incoming tasks before aggregating
+	// the tasks in a group.
+	//
+	// If unset or zero, no delay limit is used.
+	GroupMaxDelay time.Duration
+
+	// GroupMaxSize specifies the maximum number of tasks that can be aggregated into a single task within a group.
+	// If GroupMaxSize is reached, the server will aggregate the tasks into one immediately.
+	//
+	// If unset or zero, no size limit is used.
+	GroupMaxSize int
+
+	// GroupAggregator specifies the aggregation function used to aggregate multiple tasks in a group into one task.
+	//
+	// If unset or nil, the group aggregation feature will be disabled on the server.
+	GroupAggregator GroupAggregator
+}
+
+// GroupAggregator aggregates a group of tasks into one before the tasks are passed to the Handler.
+type GroupAggregator interface {
+	// Aggregate aggregates the given tasks in a group with the given group name,
+	// and returns a new task which is the aggregation of those tasks.
+	//
+	// Use NewTask(typename, payload, opts...) to set any options for the aggregated task.
+	// The Queue option, if provided, will be ignored and the aggregated task will always be enqueued
+	// to the same queue the group belonged.
+	Aggregate(group string, tasks []*Task) *Task
+}
+
+// The GroupAggregatorFunc type is an adapter to allow the use of  ordinary functions as a GroupAggregator.
+// If f is a function with the appropriate signature, GroupAggregatorFunc(f) is a GroupAggregator that calls f.
+type GroupAggregatorFunc func(group string, tasks []*Task) *Task
+
+// Aggregate calls fn(group, tasks)
+func (fn GroupAggregatorFunc) Aggregate(group string, tasks []*Task) *Task {
+	return fn(group, tasks)
 }
 
 // An ErrorHandler handles an error occured during task processing.
@@ -338,6 +385,8 @@ const (
 	defaultHealthCheckInterval = 15 * time.Second
 
 	defaultDelayedTaskCheckInterval = 5 * time.Second
+
+	defaultGroupGracePeriod = 1 * time.Minute
 )
 
 // NewServer returns a new Server given a redis connection option
@@ -387,6 +436,14 @@ func NewServer(r RedisConnOpt, cfg Config) *Server {
 	if healthcheckInterval == 0 {
 		healthcheckInterval = defaultHealthCheckInterval
 	}
+	// TODO: Create a helper to check for zero value and fall back to default (e.g. getDurationOrDefault())
+	groupGracePeriod := cfg.GroupGracePeriod
+	if groupGracePeriod == 0 {
+		groupGracePeriod = defaultGroupGracePeriod
+	}
+	if groupGracePeriod < time.Second {
+		panic("GroupGracePeriod cannot be less than a second")
+	}
 	logger := log.NewLogger(cfg.Logger)
 	loglevel := cfg.LogLevel
 	if loglevel == level_unspecified {
@@ -468,6 +525,15 @@ func NewServer(r RedisConnOpt, cfg Config) *Server {
 		queues:   qnames,
 		interval: 8 * time.Second,
 	})
+	aggregator := newAggregator(aggregatorParams{
+		logger:          logger,
+		broker:          rdb,
+		queues:          qnames,
+		gracePeriod:     groupGracePeriod,
+		maxDelay:        cfg.GroupMaxDelay,
+		maxSize:         cfg.GroupMaxSize,
+		groupAggregator: cfg.GroupAggregator,
+	})
 	return &Server{
 		logger:        logger,
 		broker:        rdb,
@@ -480,6 +546,7 @@ func NewServer(r RedisConnOpt, cfg Config) *Server {
 		recoverer:     recoverer,
 		healthchecker: healthchecker,
 		janitor:       janitor,
+		aggregator:    aggregator,
 	}
 }
 
@@ -556,6 +623,7 @@ func (srv *Server) Start(handler Handler) error {
 	srv.forwarder.start(&srv.wg)
 	srv.processor.start(&srv.wg)
 	srv.janitor.start(&srv.wg)
+	srv.aggregator.start(&srv.wg)
 	return nil
 }
 
@@ -601,6 +669,7 @@ func (srv *Server) Shutdown() {
 	srv.syncer.shutdown()
 	srv.subscriber.shutdown()
 	srv.janitor.shutdown()
+	srv.aggregator.shutdown()
 	srv.healthchecker.shutdown()
 	srv.heartbeater.shutdown()
 	srv.wg.Wait()

server_test.go

@@ -11,9 +11,9 @@ import (
 	"testing"
 	"time"
 
-	"github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/rdb"
 	"github.com/hibiken/asynq/internal/testbroker"
+	"github.com/hibiken/asynq/internal/testutil"
 	"go.uber.org/goleak"
 )
 
@@ -40,12 +40,12 @@ func TestServer(t *testing.T) {
 		t.Fatal(err)
 	}
 
-	_, err = c.Enqueue(NewTask("send_email", asynqtest.JSON(map[string]interface{}{"recipient_id": 123})))
+	_, err = c.Enqueue(NewTask("send_email", testutil.JSON(map[string]interface{}{"recipient_id": 123})))
 	if err != nil {
 		t.Errorf("could not enqueue a task: %v", err)
 	}
 
-	_, err = c.Enqueue(NewTask("send_email", asynqtest.JSON(map[string]interface{}{"recipient_id": 456})), ProcessIn(1*time.Hour))
+	_, err = c.Enqueue(NewTask("send_email", testutil.JSON(map[string]interface{}{"recipient_id": 456})), ProcessIn(1*time.Hour))
 	if err != nil {
 		t.Errorf("could not enqueue a task: %v", err)
 	}

syncer_test.go

@@ -11,9 +11,9 @@ import (
 	"testing"
 	"time"
 
-	h "github.com/hibiken/asynq/internal/asynqtest"
 	"github.com/hibiken/asynq/internal/base"
 	"github.com/hibiken/asynq/internal/rdb"
+	h "github.com/hibiken/asynq/internal/testutil"
 )
 
 func TestSyncer(t *testing.T) {

tools/asynq/cmd/group.go

@@ -0,0 +1,48 @@
+// Copyright 2022 Kentaro Hibino. All rights reserved.
+// Use of this source code is governed by a MIT license
+// that can be found in the LICENSE file.
+
+package cmd
+
+import (
+	"fmt"
+	"os"
+
+	"github.com/spf13/cobra"
+)
+
+func init() {
+	rootCmd.AddCommand(groupCmd)
+	groupCmd.AddCommand(groupListCmd)
+	groupListCmd.Flags().StringP("queue", "q", "", "queue to inspect")
+	groupListCmd.MarkFlagRequired("queue")
+}
+
+var groupCmd = &cobra.Command{
+	Use:   "group",
+	Short: "Manage groups",
+}
+
+var groupListCmd = &cobra.Command{
+	Use:   "ls",
+	Short: "List groups",
+	Args:  cobra.NoArgs,
+	Run:   groupLists,
+}
+
+func groupLists(cmd *cobra.Command, args []string) {
+	qname, err := cmd.Flags().GetString("queue")
+	if err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+	inspector := createInspector()
+	groups, err := inspector.Groups(qname)
+	if len(groups) == 0 {
+		fmt.Printf("No groups found in queue %q\n", qname)
+		return
+	}
+	for _, g := range groups {
+		fmt.Println(g.Group)
+	}
+}

tools/asynq/cmd/queue.go

@@ -145,12 +145,13 @@ func printQueueInfo(info *asynq.QueueInfo) {
 	bold.Println("Queue Info")
 	fmt.Printf("Name:   %s\n", info.Queue)
 	fmt.Printf("Size:   %d\n", info.Size)
+	fmt.Printf("Groups: %d\n", info.Groups)
 	fmt.Printf("Paused: %t\n\n", info.Paused)
 	bold.Println("Task Count by State")
 	printTable(
-		[]string{"active", "pending", "scheduled", "retry", "archived", "completed"},
+		[]string{"active", "pending", "aggregating", "scheduled", "retry", "archived", "completed"},
 		func(w io.Writer, tmpl string) {
-			fmt.Fprintf(w, tmpl, info.Active, info.Pending, info.Scheduled, info.Retry, info.Archived, info.Completed)
+			fmt.Fprintf(w, tmpl, info.Active, info.Pending, info.Aggregating, info.Scheduled, info.Retry, info.Archived, info.Completed)
 		},
 	)
 	fmt.Println()

tools/asynq/cmd/stats.go

@@ -59,15 +59,16 @@ func init() {
 }
 
 type AggregateStats struct {
-	Active    int       `json:"active"`
-	Pending   int       `json:"pending"`
-	Scheduled int       `json:"scheduled"`
-	Retry     int       `json:"retry"`
-	Archived  int       `json:"archived"`
-	Completed int       `json:"completed"`
-	Processed int       `json:"processed"`
-	Failed    int       `json:"failed"`
-	Timestamp time.Time `json:"timestamp"`
+	Active      int       `json:"active"`
+	Pending     int       `json:"pending"`
+	Aggregating int       `json:"aggregating"`
+	Scheduled   int       `json:"scheduled"`
+	Retry       int       `json:"retry"`
+	Archived    int       `json:"archived"`
+	Completed   int       `json:"completed"`
+	Processed   int       `json:"processed"`
+	Failed      int       `json:"failed"`
+	Timestamp   time.Time `json:"timestamp"`
 }
 
 type FullStats struct {
@@ -95,6 +96,7 @@ func stats(cmd *cobra.Command, args []string) {
 		}
 		aggStats.Active += s.Active
 		aggStats.Pending += s.Pending
+		aggStats.Aggregating += s.Aggregating
 		aggStats.Scheduled += s.Scheduled
 		aggStats.Retry += s.Retry
 		aggStats.Archived += s.Archived
@@ -155,13 +157,13 @@ func stats(cmd *cobra.Command, args []string) {
 }
 
 func printStatsByState(s *AggregateStats) {
-	format := strings.Repeat("%v\t", 6) + "\n"
+	format := strings.Repeat("%v\t", 7) + "\n"
 	tw := new(tabwriter.Writer).Init(os.Stdout, 0, 8, 2, ' ', 0)
-	fmt.Fprintf(tw, format, "active", "pending", "scheduled", "retry", "archived", "completed")
-	width := maxInt(9 /* defaultWidth */, maxWidthOf(s.Active, s.Pending, s.Scheduled, s.Retry, s.Archived, s.Completed)) // length of widest column
+	fmt.Fprintf(tw, format, "active", "pending", "aggregating", "scheduled", "retry", "archived", "completed")
+	width := maxInt(9 /* defaultWidth */, maxWidthOf(s.Active, s.Pending, s.Aggregating, s.Scheduled, s.Retry, s.Archived, s.Completed)) // length of widest column
 	sep := strings.Repeat("-", width)
-	fmt.Fprintf(tw, format, sep, sep, sep, sep, sep, sep)
-	fmt.Fprintf(tw, format, s.Active, s.Pending, s.Scheduled, s.Retry, s.Archived, s.Completed)
+	fmt.Fprintf(tw, format, sep, sep, sep, sep, sep, sep, sep)
+	fmt.Fprintf(tw, format, s.Active, s.Pending, s.Aggregating, s.Scheduled, s.Retry, s.Archived, s.Completed)
 	tw.Flush()
 }
 

tools/asynq/cmd/task.go

@@ -22,6 +22,7 @@ func init() {
 	taskListCmd.Flags().StringP("state", "s", "", "state of the tasks to inspect")
 	taskListCmd.Flags().Int("page", 1, "page number")
 	taskListCmd.Flags().Int("size", 30, "page size")
+	taskListCmd.Flags().StringP("group", "g", "", "group to inspect (required for listing aggregating tasks)")
 	taskListCmd.MarkFlagRequired("queue")
 	taskListCmd.MarkFlagRequired("state")
 
@@ -83,6 +84,7 @@ var taskListCmd = &cobra.Command{
 The value for the state flag should be one of:
 - active
 - pending
+- aggregating
 - scheduled
 - retry
 - archived
@@ -92,12 +94,19 @@ List opeartion paginates the result set.
 By default, the command fetches the first 30 tasks.
 Use --page and --size flags to specify the page number and size.
 
+
 Example: 
 To list pending tasks from "default" queue, run
   asynq task ls --queue=default --state=pending
 
 To list the tasks from the second page, run
-  asynq task ls --queue=default --state=pending --page=1`,
+  asynq task ls --queue=default --state=pending --page=1
+
+For aggregating tasks, additional --group flag is required.
+
+Example:
+  asynq task ls --queue=default --state=aggregating --group=mygroup
+`,
 	Run: taskList,
 }
 
@@ -192,6 +201,17 @@ func taskList(cmd *cobra.Command, args []string) {
 		listArchivedTasks(qname, pageNum, pageSize)
 	case "completed":
 		listCompletedTasks(qname, pageNum, pageSize)
+	case "aggregating":
+		group, err := cmd.Flags().GetString("group")
+		if err != nil {
+			fmt.Println(err)
+			os.Exit(1)
+		}
+		if group == "" {
+			fmt.Println("Flag --group is required for listing aggregating tasks")
+			os.Exit(1)
+		}
+		listAggregatingTasks(qname, group, pageNum, pageSize)
 	default:
 		fmt.Printf("error: state=%q is not supported\n", state)
 		os.Exit(1)
@@ -334,6 +354,27 @@ func listCompletedTasks(qname string, pageNum, pageSize int) {
 		})
 }
 
+func listAggregatingTasks(qname, group string, pageNum, pageSize int) {
+	i := createInspector()
+	tasks, err := i.ListAggregatingTasks(qname, group, asynq.PageSize(pageSize), asynq.Page(pageNum))
+	if err != nil {
+		fmt.Println(err)
+		os.Exit(1)
+	}
+	if len(tasks) == 0 {
+		fmt.Printf("No aggregating tasks in group %q \n", group)
+		return
+	}
+	printTable(
+		[]string{"ID", "Type", "Payload", "Group"},
+		func(w io.Writer, tmpl string) {
+			for _, t := range tasks {
+				fmt.Fprintf(w, tmpl, t.ID, t.Type, sprintBytes(t.Payload), t.Group)
+			}
+		},
+	)
+}
+
 func taskCancel(cmd *cobra.Command, args []string) {
 	i := createInspector()
 	for _, id := range args {

tools/go.mod

@@ -13,3 +13,5 @@ require (
 	github.com/spf13/cobra v1.1.1
 	github.com/spf13/viper v1.7.0
 )
+
+replace github.com/hibiken/asynq => ../