[#754] blobstor: Estimate compressability

Now it is possible to enable compressability estimation.
If data is likely uncompressable, it should reduce CPU time and memory.

Signed-off-by: Dmitrii Stepanov <d.stepanov@yadro.com>

cmd/frostfs-node/config.go

@@ -105,7 +105,10 @@ type applicationConfiguration struct {
 }
 
 type shardCfg struct {
-	compress                  bool
+	compress                         bool
+	estimateCompressibility          bool
+	estimateCompressibilityThreshold float64
+
 	smallSizeObjectLimit      uint64
 	uncompressableContentType []string
 	refillMetabase            bool
@@ -217,6 +220,8 @@ func (a *applicationConfiguration) updateShardConfig(c *config.Config, oldConfig
 	newConfig.refillMetabase = oldConfig.RefillMetabase()
 	newConfig.mode = oldConfig.Mode()
 	newConfig.compress = oldConfig.Compress()
+	newConfig.estimateCompressibility = oldConfig.EstimateCompressibility()
+	newConfig.estimateCompressibilityThreshold = oldConfig.EstimateCompressibilityThreshold()
 	newConfig.uncompressableContentType = oldConfig.UncompressableContentTypes()
 	newConfig.smallSizeObjectLimit = oldConfig.SmallSizeLimit()
 
@@ -830,6 +835,8 @@ func (c *cfg) getShardOpts(shCfg shardCfg) shardOptsWithID {
 	blobstoreOpts := []blobstor.Option{
 		blobstor.WithCompressObjects(shCfg.compress),
 		blobstor.WithUncompressableContentTypes(shCfg.uncompressableContentType),
+		blobstor.WithCompressibilityEstimate(shCfg.estimateCompressibility),
+		blobstor.WithCompressibilityEstimateThreshold(shCfg.estimateCompressibilityThreshold),
 		blobstor.WithStorages(ss),
 		blobstor.WithLogger(c.log),
 	}

cmd/frostfs-node/config/cast.go

@@ -223,3 +223,15 @@ func parseSizeInBytes(sizeStr string) uint64 {
 	size := cast.ToFloat64(sizeStr)
 	return safeMul(size, multiplier)
 }
+
+// FloatOrDefault reads a configuration value
+// from c by name and casts it to float64.
+//
+// Returns defaultValue if the value can not be casted.
+func FloatOrDefault(c *Config, name string, defaultValue float64) float64 {
+	v, err := cast.ToFloat64E(c.Value(name))
+	if err != nil {
+		return defaultValue
+	}
+	return v
+}

cmd/frostfs-node/config/engine/config_test.go

@@ -84,6 +84,8 @@ func TestEngineSection(t *testing.T) {
 
 				require.Equal(t, true, sc.Compress())
 				require.Equal(t, []string{"audio/*", "video/*"}, sc.UncompressableContentTypes())
+				require.Equal(t, true, sc.EstimateCompressibility())
+				require.Equal(t, float64(0.7), sc.EstimateCompressibilityThreshold())
 				require.EqualValues(t, 102400, sc.SmallSizeLimit())
 
 				require.Equal(t, 2, len(ss))

cmd/frostfs-node/config/engine/shard/config.go

@@ -16,8 +16,11 @@ import (
 // which provides access to Shard configurations.
 type Config config.Config
 
-// SmallSizeLimitDefault is a default limit of small objects payload in bytes.
-const SmallSizeLimitDefault = 1 << 20
+const (
+	// SmallSizeLimitDefault is a default limit of small objects payload in bytes.
+	SmallSizeLimitDefault                   = 1 << 20
+	EstimateCompressibilityThresholdDefault = 0.1
+)
 
 // From wraps config section into Config.
 func From(c *config.Config) *Config {
@@ -43,6 +46,30 @@ func (x *Config) UncompressableContentTypes() []string {
 		"compression_exclude_content_types")
 }
 
+// EstimateCompressibility returns the value of "estimate_compressibility" config parameter.
+//
+// Returns false if the value is not a valid bool.
+func (x *Config) EstimateCompressibility() bool {
+	return config.BoolSafe(
+		(*config.Config)(x),
+		"compression_estimate_compressibility",
+	)
+}
+
+// EstimateCompressibilityThreshold returns the value of "estimate_compressibility_threshold" config parameter.
+//
+// Returns EstimateCompressibilityThresholdDefault if the value is not defined, not valid float or not in range [0.0; 1.0].
+func (x *Config) EstimateCompressibilityThreshold() float64 {
+	v := config.FloatOrDefault(
+		(*config.Config)(x),
+		"compression_estimate_compressibility_threshold",
+		EstimateCompressibilityThresholdDefault)
+	if v < 0.0 || v > 1.0 {
+		return EstimateCompressibilityThresholdDefault
+	}
+	return v
+}
+
 // SmallSizeLimit returns the value of "small_object_size" config parameter.
 //
 // Returns SmallSizeLimitDefault if the value is not a positive number.

config/example/node.env

@@ -113,6 +113,8 @@ FROSTFS_STORAGE_SHARD_0_METABASE_MAX_BATCH_DELAY=10ms
 ### Blobstor config
 FROSTFS_STORAGE_SHARD_0_COMPRESS=true
 FROSTFS_STORAGE_SHARD_0_COMPRESSION_EXCLUDE_CONTENT_TYPES="audio/* video/*"
+FROSTFS_STORAGE_SHARD_0_COMPRESSION_ESTIMATE_COMPRESSIBILITY=true
+FROSTFS_STORAGE_SHARD_0_COMPRESSION_ESTIMATE_COMPRESSIBILITY_THRESHOLD=0.7
 FROSTFS_STORAGE_SHARD_0_SMALL_OBJECT_SIZE=102400
 ### Blobovnicza config
 FROSTFS_STORAGE_SHARD_0_BLOBSTOR_0_PATH=tmp/0/blob/blobovnicza

config/example/node.json

@@ -160,6 +160,8 @@
         "compression_exclude_content_types": [
           "audio/*", "video/*"
         ],
+        "compression_estimate_compressibility": true,
+        "compression_estimate_compressibility_threshold": 0.7,
         "small_object_size": 102400,
         "blobstor": [
           {

config/example/node.yaml

@@ -178,6 +178,8 @@ storage:
       compression_exclude_content_types:
         - audio/*
         - video/*
+      compression_estimate_compressibility: true
+      compression_estimate_compressibility_threshold: 0.7
 
       blobstor:
         - type: blobovnicza

docs/storage-node-configuration.md

@@ -179,15 +179,17 @@ The following table describes configuration for each shard.
 
 | Parameter                           | Type                                        | Default value | Description                                                                                                                                                                                                       |
 |-------------------------------------|---------------------------------------------|---------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| `compress`                          | `bool`                                      | `false`       | Flag to enable compression.                                                                                                                                                                                       |
-| `compression_exclude_content_types` | `[]string`                                  |               | List of content-types to disable compression for. Content-type is taken from `Content-Type` object attribute. Each element can contain a star `*` as a first (last) character, which matches any prefix (suffix). |
-| `mode`                              | `string`                                    | `read-write`  | Shard Mode.<br/>Possible values:  `read-write`, `read-only`, `degraded`, `degraded-read-only`, `disabled`                                                                                                         |
-| `resync_metabase`                   | `bool`                                      | `false`       | Flag to enable metabase resync on start.                                                                                                                                                                          |
-| `writecache`                        | [Writecache config](#writecache-subsection) |               | Write-cache configuration.                                                                                                                                                                                        |
-| `metabase`                          | [Metabase config](#metabase-subsection)     |               | Metabase configuration.                                                                                                                                                                                           |
-| `blobstor`                          | [Blobstor config](#blobstor-subsection)     |               | Blobstor configuration.                                                                                                                                                                                           |
-| `small_object_size`                 | `size`                                      | `1M`          | Maximum size of an object stored in blobovnicza tree.                                                                                                                                                             |
-| `gc`                                | [GC config](#gc-subsection)                 |               | GC configuration.                                                                                                                                                                                                 |
+| `compress`                                       | `bool`                                      | `false`       | Flag to enable compression.                                                                                                                                                                                       |
+| `compression_exclude_content_types`              | `[]string`                                  |               | List of content-types to disable compression for. Content-type is taken from `Content-Type` object attribute. Each element can contain a star `*` as a first (last) character, which matches any prefix (suffix). |
+| `compression_estimate_compressibility`           | `bool`                                      |  `false`      | If `true`, then noramalized compressibility estimation is used to decide compress data or not.  |
+| `compression_estimate_compressibility_threshold` | `float`                                     |   `0.1`       | Normilized compressibility estimate threshold: data will compress if estimation if greater than this value.  |
+| `mode`                                           | `string`                                    | `read-write`  | Shard Mode.<br/>Possible values:  `read-write`, `read-only`, `degraded`, `degraded-read-only`, `disabled`                                                                                                         |
+| `resync_metabase`                                | `bool`                                      | `false`       | Flag to enable metabase resync on start.                                                                                                                                                                          |
+| `writecache`                                     | [Writecache config](#writecache-subsection) |               | Write-cache configuration.                                                                                                                                                                                        |
+| `metabase`                                       | [Metabase config](#metabase-subsection)     |               | Metabase configuration.                                                                                                                                                                                           |
+| `blobstor`                                       | [Blobstor config](#blobstor-subsection)     |               | Blobstor configuration.                                                                                                                                                                                           |
+| `small_object_size`                              | `size`                                      | `1M`          | Maximum size of an object stored in blobovnicza tree.                                                                                                                                                             |
+| `gc`                                             | [GC config](#gc-subsection)                 |               | GC configuration.                                                                                                                                                                                                 |
 
 ### `blobstor` subsection
 

pkg/local_object_storage/blobstor/blobstor.go

@@ -107,6 +107,27 @@ func WithCompressObjects(comp bool) Option {
 	}
 }
 
+// WithCompressibilityEstimate returns an option to use
+// normilized compressibility estimate to decide compress
+// data or not.
+//
+// See https://github.com/klauspost/compress/blob/v1.17.2/compressible.go#L5
+func WithCompressibilityEstimate(v bool) Option {
+	return func(c *cfg) {
+		c.compression.UseCompressEstimation = v
+	}
+}
+
+// WithCompressibilityEstimateThreshold returns an option to set
+// normilized compressibility estimate threshold.
+//
+// See https://github.com/klauspost/compress/blob/v1.17.2/compressible.go#L5
+func WithCompressibilityEstimateThreshold(threshold float64) Option {
+	return func(c *cfg) {
+		c.compression.CompressEstimationThreshold = threshold
+	}
+}
+
 // WithUncompressableContentTypes returns option to disable decompression
 // for specific content types as seen by object.AttributeContentType attribute.
 func WithUncompressableContentTypes(values []string) Option {

pkg/local_object_storage/blobstor/compression/bench_test.go

@@ -3,8 +3,10 @@ package compression
 import (
 	"crypto/rand"
 	"fmt"
+	"log"
 	"testing"
 
+	"github.com/klauspost/compress"
 	"github.com/stretchr/testify/require"
 )
 
@@ -47,3 +49,50 @@ func notSoRandomSlice(size, blockSize int) []byte {
 	}
 	return data
 }
+
+func BenchmarkCompressionRealVSEstimate(b *testing.B) {
+	var total float64 // to prevent from compiler optimizations
+	maxSize := 60 * 1024 * 1024
+	b.Run("estimate", func(b *testing.B) {
+		b.ResetTimer()
+
+		c := &Config{
+			Enabled: true,
+		}
+		require.NoError(b, c.Init())
+
+		for size := 1024; size <= maxSize; size *= 2 {
+			data := make([]byte, size)
+			_, err := rand.Reader.Read(data)
+			require.NoError(b, err)
+
+			b.StartTimer()
+			estimation := compress.Estimate(data)
+			total += estimation
+			b.StopTimer()
+		}
+	})
+
+	b.Run("compress", func(b *testing.B) {
+		b.ResetTimer()
+
+		c := &Config{
+			Enabled: true,
+		}
+		require.NoError(b, c.Init())
+
+		for size := 1024; size <= maxSize; size *= 2 {
+			data := make([]byte, size)
+			_, err := rand.Reader.Read(data)
+			require.NoError(b, err)
+
+			b.StartTimer()
+			maxSize := c.encoder.MaxEncodedSize(len(data))
+			compressed := c.encoder.EncodeAll(data, make([]byte, 0, maxSize))
+			total += float64(len(compressed)) / float64(len(data))
+			b.StopTimer()
+		}
+	})
+
+	log.Println(total)
+}

pkg/local_object_storage/blobstor/compression/compress.go

@@ -5,6 +5,7 @@ import (
 	"strings"
 
 	objectSDK "git.frostfs.info/TrueCloudLab/frostfs-sdk-go/object"
+	"github.com/klauspost/compress"
 	"github.com/klauspost/compress/zstd"
 )
 
@@ -13,6 +14,9 @@ type Config struct {
 	Enabled                    bool
 	UncompressableContentTypes []string
 
+	UseCompressEstimation       bool
+	CompressEstimationThreshold float64
+
 	encoder *zstd.Encoder
 	decoder *zstd.Decoder
 }
@@ -82,6 +86,17 @@ func (c *Config) Compress(data []byte) []byte {
 	if c == nil || !c.Enabled {
 		return data
 	}
+	if c.UseCompressEstimation {
+		estimated := compress.Estimate(data)
+		if estimated >= c.CompressEstimationThreshold {
+			return c.compress(data)
+		}
+		return data
+	}
+	return c.compress(data)
+}
+
+func (c *Config) compress(data []byte) []byte {
 	maxSize := c.encoder.MaxEncodedSize(len(data))
 	compressed := c.encoder.EncodeAll(data, make([]byte, 0, maxSize))
 	if len(data) < len(compressed) {