Fast, concurrent, evicting in-memory cache written to keep big number of entries without impact on performance. BigCache keeps entries on heap but omits GC for them. To achieve that operations on bytes arrays take place, therefore entries (de)serialization in front of the cache will be needed in most use cases.

import "github.com/allegro/bigcache"

cache, _ := bigcache.NewBigCache(bigcache.DefaultConfig(10 * time.Minute))

cache.Set("my-unique-key", []byte("value"))

entry, _ := cache.Get("my-unique-key")
fmt.Println(string(entry))

When cache load can be predicted in advance then it is better to use custom initialization because additional memory allocation can be avoided in that way.

import (
	"log"

	"github.com/allegro/bigcache"
)

config := bigcache.Config {
		// number of shards (must be a power of 2)
		Shards: 1024,
		// time after which entry can be evicted
		LifeWindow: 10 * time.Minute,
		// rps * lifeWindow, used only in initial memory allocation
		MaxEntriesInWindow: 1000 * 10 * 60,
		// max entry size in bytes, used only in initial memory allocation
		MaxEntrySize: 500,
		// prints information about additional memory allocation
		Verbose: true,
		// cache will not allocate more memory than this limit, value in MB
		// if value is reached then the oldest entries can be overridden for the new ones
		// 0 value means no size limit
		HardMaxCacheSize: 8192,
		// callback fired when the oldest entry is removed because of its
		// expiration time or no space left for the new entry. Default value is nil which
		// means no callback and it prevents from unwrapping the oldest entry.
		OnRemove: nil,
	}

cache, initErr := bigcache.NewBigCache(config)
if initErr != nil {
	log.Fatal(initErr)
}

cache.Set("my-unique-key", []byte("value"))

if entry, err := cache.Get("my-unique-key"); err == nil {
	fmt.Println(string(entry))
}

Three caches were compared: bigcache, freecache and map. Benchmark tests were made using an i7-6700K with 32GB of RAM on Windows 10.

cd caches_bench; go test -bench=. -benchtime=10s ./... -timeout 30m

BenchmarkMapSet-8                     	 2000000	       716 ns/op	     336 B/op	       3 allocs/op
BenchmarkConcurrentMapSet-8           	 1000000	      1292 ns/op	     347 B/op	       8 allocs/op
BenchmarkFreeCacheSet-8               	 3000000	       501 ns/op	     371 B/op	       3 allocs/op
BenchmarkBigCacheSet-8                	 3000000	       482 ns/op	     303 B/op	       2 allocs/op
BenchmarkMapGet-8                     	 5000000	       309 ns/op	      24 B/op	       1 allocs/op
BenchmarkConcurrentMapGet-8           	 2000000	       659 ns/op	      24 B/op	       2 allocs/op
BenchmarkFreeCacheGet-8               	 3000000	       541 ns/op	     152 B/op	       3 allocs/op
BenchmarkBigCacheGet-8                	 3000000	       420 ns/op	     152 B/op	       3 allocs/op
BenchmarkBigCacheSetParallel-8        	10000000	       184 ns/op	     313 B/op	       3 allocs/op
BenchmarkFreeCacheSetParallel-8       	10000000	       195 ns/op	     357 B/op	       4 allocs/op
BenchmarkConcurrentMapSetParallel-8   	 5000000	       242 ns/op	     200 B/op	       6 allocs/op
BenchmarkBigCacheGetParallel-8        	20000000	       100 ns/op	     152 B/op	       4 allocs/op
BenchmarkFreeCacheGetParallel-8       	10000000	       133 ns/op	     152 B/op	       4 allocs/op
BenchmarkConcurrentMapGetParallel-8   	10000000	       202 ns/op	      24 B/op	       2 allocs/op

Writes and reads in bigcache are faster than in freecache. Writes to map are the slowest.

cd caches_bench; go run caches_gc_overhead_comparison.go

Number of entries:  20000000
GC pause for bigcache:  27.81671ms
GC pause for freecache:  30.218371ms
GC pause for map:  11.590772251s

Test shows how long are the GC pauses for caches filled with 20mln of entries. Bigcache and freecache have very similar GC pause time. It is clear that both reduce GC overhead in contrast to map which GC pause time took more than 10 seconds.

BigCache relies on optimization presented in 1.5 version of Go (issue-9477). This optimization states that if map without pointers in keys and values is used then GC will omit it’s content. Therefore BigCache uses map[uint64]uint32 where keys are hashed and values are offsets of entries.

Entries are kept in bytes array, to omit GC again. Bytes array size can grow to gigabytes without impact on performance because GC will only see single pointer to it.

Both caches provide the same core features but they reduce GC overhead in different ways. Bigcache relies on map[uint64]uint32, freecache implements its own mapping built on slices to reduce number of pointers.

Results from benchmark tests are presented above. One of the advantage of bigcache over freecache is that you don’t need to know the size of the cache in advance, because when bigcache is full, it can allocate additional memory for new entries instead of overwriting existing ones as freecache does currently. However hard max size in bigcache also can be set, check HardMaxCacheSize.

Bigcache genesis is described in allegro.tech blog post: writing a very fast cache service in Go

BigCache is released under the Apache 2.0 license (see LICENSE)