Fjord: Add Brotli channel compression support (#10358)

* wip

* wip

* fix

* fix

* fix

* fix

* address some of the bots comments

* use version bit of 1

* fix lint

* adding compression type

* update batch reader

* abstract span channel compressor

* test and singular batch compressor

* fix

* lint

* move channel compressor as interface

* add base class

* fix go mod

* test fixes

* address comments

* fix

* fix

* revert channel builder test

* revert ratio compressor test

* add checks to accept brotli only post fjord

* revemo unnecessary in test

* fix forge-std

* gofmt

* address comments

* remove methods in compressor

* fix error msg

* add compression algo flag to optional flags

* add Clone() function

---------
Co-authored-by: Roberto Bayardo <roberto.bayardo@coinbase.com>

go.mod

@@ -4,6 +4,8 @@ go 1.21
 
 require (
 	github.com/BurntSushi/toml v1.3.2
+	github.com/DataDog/zstd v1.5.2
+	github.com/andybalholm/brotli v1.1.0
 	github.com/btcsuite/btcd v0.24.0
 	github.com/btcsuite/btcd/chaincfg/chainhash v1.1.0
 	github.com/cockroachdb/pebble v0.0.0-20231018212520-f6cde3fc2fa4
@@ -55,7 +57,6 @@ require (
 )
 
 require (
-	github.com/DataDog/zstd v1.5.2 // indirect
 	github.com/Microsoft/go-winio v0.6.1 // indirect
 	github.com/VictoriaMetrics/fastcache v1.12.1 // indirect
 	github.com/allegro/bigcache v1.2.1 // indirect

go.sum

@@ -30,6 +30,8 @@ github.com/alecthomas/units v0.0.0-20190717042225-c3de453c63f4/go.mod h1:ybxpYRF
 github.com/allegro/bigcache v1.2.1-0.20190218064605-e24eb225f156/go.mod h1:Cb/ax3seSYIx7SuZdm2G2xzfwmv3TPSk2ucNfQESPXM=
 github.com/allegro/bigcache v1.2.1 h1:hg1sY1raCwic3Vnsvje6TT7/pnZba83LeFck5NrFKSc=
 github.com/allegro/bigcache v1.2.1/go.mod h1:Cb/ax3seSYIx7SuZdm2G2xzfwmv3TPSk2ucNfQESPXM=
+github.com/andybalholm/brotli v1.1.0 h1:eLKJA0d02Lf0mVpIDgYnqXcUn0GqVmEFny3VuID1U3M=
+github.com/andybalholm/brotli v1.1.0/go.mod h1:sms7XGricyQI9K10gOSf56VKKWS4oLer58Q+mhRPtnY=
 github.com/anmitsu/go-shlex v0.0.0-20161002113705-648efa622239/go.mod h1:2FmKhYUyUczH0OGQWaF5ceTx0UBShxjsH6f8oGKYe2c=
 github.com/armon/go-metrics v0.0.0-20190430140413-ec5e00d3c878/go.mod h1:3AMJUQhVx52RsWOnlkpikZr01T/yAVN2gn0861vByNg=
 github.com/armon/go-metrics v0.3.8/go.mod h1:4O98XIr/9W0sxpJ8UaYkvjk10Iff7SnFrb4QAOwNTFc=

op-batcher/batcher/channel_builder.go

@@ -86,7 +86,7 @@ func NewChannelBuilder(cfg ChannelConfig, rollupCfg rollup.Config, latestL1Origi
 	}
 	var co derive.ChannelOut
 	if cfg.BatchType == derive.SpanBatchType {
-		co, err = derive.NewSpanChannelOut(rollupCfg.Genesis.L2Time, rollupCfg.L2ChainID, cfg.CompressorConfig.TargetOutputSize)
+		co, err = derive.NewSpanChannelOut(rollupCfg.Genesis.L2Time, rollupCfg.L2ChainID, cfg.CompressorConfig.TargetOutputSize, cfg.CompressorConfig.CompressionAlgo)
 	} else {
 		co, err = derive.NewSingularChannelOut(c)
 	}

op-batcher/batcher/channel_builder_test.go

@@ -297,7 +297,6 @@ func TestChannelBuilderBatchType(t *testing.T) {
 		{"ChannelBuilder_PendingFrames_TotalFrames", ChannelBuilder_PendingFrames_TotalFrames},
 		{"ChannelBuilder_InputBytes", ChannelBuilder_InputBytes},
 		{"ChannelBuilder_OutputBytes", ChannelBuilder_OutputBytes},
-		{"ChannelBuilder_OutputWrongFramePanic", ChannelBuilder_OutputWrongFramePanic},
 	}
 	for _, test := range tests {
 		test := test
@@ -413,7 +412,8 @@ func TestChannelBuilder_OutputFrames(t *testing.T) {
 
 	// Check how many ready bytes
 	require.Greater(t, uint64(cb.co.ReadyBytes()+derive.FrameV0OverHeadSize), channelConfig.MaxFrameSize)
-	require.Equal(t, 0, cb.PendingFrames())
+
+	require.Equal(t, 0, cb.PendingFrames()) // always 0 because non compressor
 
 	// The channel should not be full
 	// but we want to output the frames for testing anyways
@@ -430,11 +430,27 @@ func TestChannelBuilder_OutputFrames(t *testing.T) {
 }
 
 func TestChannelBuilder_OutputFrames_SpanBatch(t *testing.T) {
+	for _, algo := range derive.CompressionAlgoTypes {
+		t.Run("ChannelBuilder_OutputFrames_SpanBatch_"+algo.String(), func(t *testing.T) {
+			if algo.IsBrotli() {
+				ChannelBuilder_OutputFrames_SpanBatch(t, algo) // to fill faster for brotli
+			} else {
+				ChannelBuilder_OutputFrames_SpanBatch(t, algo)
+			}
+		})
+	}
+}
+
+func ChannelBuilder_OutputFrames_SpanBatch(t *testing.T, algo derive.CompressionAlgo) {
 	channelConfig := defaultTestChannelConfig()
 	channelConfig.MaxFrameSize = 20 + derive.FrameV0OverHeadSize
+	if algo.IsBrotli() {
+		channelConfig.TargetNumFrames = 3
+	} else {
 		channelConfig.TargetNumFrames = 5
+	}
 	channelConfig.BatchType = derive.SpanBatchType
-	channelConfig.InitRatioCompressor(1)
+	channelConfig.InitRatioCompressor(1, algo)
 
 	// Construct the channel builder
 	cb, err := NewChannelBuilder(channelConfig, defaultTestRollupConfig, latestL1BlockOrigin)
@@ -453,6 +469,10 @@ func TestChannelBuilder_OutputFrames_SpanBatch(t *testing.T) {
 	for {
 		err = addMiniBlock(cb)
 		if err == nil {
+			if cb.IsFull() {
+				// this happens when the data exactly fills the channel
+				break
+			}
 			require.False(t, cb.IsFull())
 			// There should be no ready bytes until the channel is full
 			require.Equal(t, cb.co.ReadyBytes(), 0)
@@ -504,7 +524,7 @@ func ChannelBuilder_OutputFramesMaxFrameIndex(t *testing.T, batchType uint) {
 	channelConfig := defaultTestChannelConfig()
 	channelConfig.MaxFrameSize = derive.FrameV0OverHeadSize + 1
 	channelConfig.TargetNumFrames = math.MaxUint16 + 1
-	channelConfig.InitRatioCompressor(.1)
+	channelConfig.InitRatioCompressor(.1, derive.Zlib)
 	channelConfig.BatchType = batchType
 
 	rng := rand.New(rand.NewSource(123))
@@ -546,8 +566,8 @@ func TestChannelBuilder_FullShadowCompressor(t *testing.T) {
 		TargetNumFrames: 1,
 		BatchType:       derive.SpanBatchType,
 	}
-	cfg.InitShadowCompressor()
 
+	cfg.InitShadowCompressor(derive.Zlib)
 	cb, err := NewChannelBuilder(cfg, defaultTestRollupConfig, latestL1BlockOrigin)
 	require.NoError(err)
 
@@ -577,7 +597,7 @@ func ChannelBuilder_AddBlock(t *testing.T, batchType uint) {
 	channelConfig.MaxFrameSize = 20 + derive.FrameV0OverHeadSize
 	channelConfig.TargetNumFrames = 2
 	// Configure the Input Threshold params so we observe a full channel
-	channelConfig.InitRatioCompressor(1)
+	channelConfig.InitRatioCompressor(1, derive.Zlib)
 
 	// Construct the channel builder
 	cb, err := NewChannelBuilder(channelConfig, defaultTestRollupConfig, latestL1BlockOrigin)
@@ -700,7 +720,7 @@ func ChannelBuilder_PendingFrames_TotalFrames(t *testing.T, batchType uint) {
 	cfg.MaxFrameSize = 1000
 	cfg.TargetNumFrames = tnf
 	cfg.BatchType = batchType
-	cfg.InitShadowCompressor()
+	cfg.InitShadowCompressor(derive.Zlib)
 	cb, err := NewChannelBuilder(cfg, defaultTestRollupConfig, latestL1BlockOrigin)
 	require.NoError(err)
 
@@ -782,7 +802,7 @@ func ChannelBuilder_OutputBytes(t *testing.T, batchType uint) {
 	cfg.MaxFrameSize = 1000
 	cfg.TargetNumFrames = 16
 	cfg.BatchType = batchType
-	cfg.InitRatioCompressor(1.0)
+	cfg.InitRatioCompressor(1.0, derive.Zlib)
 	cb, err := NewChannelBuilder(cfg, defaultTestRollupConfig, latestL1BlockOrigin)
 	require.NoError(err, "NewChannelBuilder")
 

op-batcher/batcher/channel_config.go

@@ -53,25 +53,26 @@ type ChannelConfig struct {
 // value consistent with cc.TargetNumFrames and cc.MaxFrameSize.
 // comprKind can be the empty string, in which case the default compressor will
 // be used.
-func (cc *ChannelConfig) InitCompressorConfig(approxComprRatio float64, comprKind string) {
+func (cc *ChannelConfig) InitCompressorConfig(approxComprRatio float64, comprKind string, compressionAlgo derive.CompressionAlgo) {
 	cc.CompressorConfig = compressor.Config{
 		// Compressor output size needs to account for frame encoding overhead
 		TargetOutputSize: MaxDataSize(cc.TargetNumFrames, cc.MaxFrameSize),
 		ApproxComprRatio: approxComprRatio,
 		Kind:             comprKind,
+		CompressionAlgo:  compressionAlgo,
 	}
 }
 
-func (cc *ChannelConfig) InitRatioCompressor(approxComprRatio float64) {
-	cc.InitCompressorConfig(approxComprRatio, compressor.RatioKind)
+func (cc *ChannelConfig) InitRatioCompressor(approxComprRatio float64, compressionAlgo derive.CompressionAlgo) {
+	cc.InitCompressorConfig(approxComprRatio, compressor.RatioKind, compressionAlgo)
 }
 
-func (cc *ChannelConfig) InitShadowCompressor() {
-	cc.InitCompressorConfig(0, compressor.ShadowKind)
+func (cc *ChannelConfig) InitShadowCompressor(compressionAlgo derive.CompressionAlgo) {
+	cc.InitCompressorConfig(0, compressor.ShadowKind, compressionAlgo)
 }
 
 func (cc *ChannelConfig) InitNoneCompressor() {
-	cc.InitCompressorConfig(0, compressor.NoneKind)
+	cc.InitCompressorConfig(0, compressor.NoneKind, derive.Zlib)
 }
 
 func (cc *ChannelConfig) MaxFramesPerTx() int {

op-batcher/batcher/channel_config_test.go

@@ -20,7 +20,7 @@ func defaultTestChannelConfig() ChannelConfig {
 		TargetNumFrames:    1,
 		BatchType:          derive.SingularBatchType,
 	}
-	c.InitRatioCompressor(0.4)
+	c.InitRatioCompressor(0.4, derive.Zlib)
 	return c
 }
 

op-batcher/batcher/channel_manager_test.go

@@ -25,7 +25,7 @@ func channelManagerTestConfig(maxFrameSize uint64, batchType uint) ChannelConfig
 		TargetNumFrames: 1,
 		BatchType:       batchType,
 	}
-	cfg.InitRatioCompressor(1)
+	cfg.InitRatioCompressor(1, derive.Zlib)
 	return cfg
 }
 
@@ -123,7 +123,7 @@ func ChannelManager_Clear(t *testing.T, batchType uint) {
 	// channels on confirmation. This would result in [TxConfirmed]
 	// clearing confirmed transactions, and resetting the pendingChannels map
 	cfg.ChannelTimeout = 10
-	cfg.InitRatioCompressor(1)
+	cfg.InitRatioCompressor(1, derive.Zlib)
 	m := NewChannelManager(log, metrics.NoopMetrics, cfg, &defaultTestRollupConfig)
 
 	// Channel Manager state should be empty by default

op-batcher/batcher/channel_test.go

@@ -4,6 +4,7 @@ import (
 	"io"
 	"testing"
 
+	"github.com/ethereum-optimism/optimism/op-batcher/compressor"
 	"github.com/ethereum-optimism/optimism/op-batcher/metrics"
 	"github.com/ethereum-optimism/optimism/op-node/rollup"
 	"github.com/ethereum-optimism/optimism/op-node/rollup/derive"
@@ -29,6 +30,9 @@ func TestChannelTimeout(t *testing.T) {
 	log := testlog.Logger(t, log.LevelCrit)
 	m := NewChannelManager(log, metrics.NoopMetrics, ChannelConfig{
 		ChannelTimeout: 100,
+		CompressorConfig: compressor.Config{
+			CompressionAlgo: derive.Zlib,
+		},
 	}, &rollup.Config{})
 	m.Clear(eth.BlockID{})
 
@@ -71,7 +75,9 @@ func TestChannelTimeout(t *testing.T) {
 // TestChannelManager_NextTxData tests the nextTxData function.
 func TestChannelManager_NextTxData(t *testing.T) {
 	log := testlog.Logger(t, log.LevelCrit)
-	m := NewChannelManager(log, metrics.NoopMetrics, ChannelConfig{}, &rollup.Config{})
+	m := NewChannelManager(log, metrics.NoopMetrics, ChannelConfig{CompressorConfig: compressor.Config{
+		CompressionAlgo: derive.Zlib,
+	}}, &rollup.Config{})
 	m.Clear(eth.BlockID{})
 
 	// Nil pending channel should return EOF
@@ -118,6 +124,9 @@ func TestChannel_NextTxData_singleFrameTx(t *testing.T) {
 	ch, err := newChannel(lgr, metrics.NoopMetrics, ChannelConfig{
 		MultiFrameTxs:   false,
 		TargetNumFrames: n,
+		CompressorConfig: compressor.Config{
+			CompressionAlgo: derive.Zlib,
+		},
 	}, &rollup.Config{}, latestL1BlockOrigin)
 	require.NoError(err)
 	chID := ch.ID()
@@ -156,6 +165,9 @@ func TestChannel_NextTxData_multiFrameTx(t *testing.T) {
 	ch, err := newChannel(lgr, metrics.NoopMetrics, ChannelConfig{
 		MultiFrameTxs:   true,
 		TargetNumFrames: n,
+		CompressorConfig: compressor.Config{
+			CompressionAlgo: derive.Zlib,
+		},
 	}, &rollup.Config{}, latestL1BlockOrigin)
 	require.NoError(err)
 	chID := ch.ID()
@@ -202,6 +214,9 @@ func TestChannelTxConfirmed(t *testing.T) {
 		// channels on confirmation. This would result in [TxConfirmed]
 		// clearing confirmed transactions, and resetting the pendingChannels map
 		ChannelTimeout: 10,
+		CompressorConfig: compressor.Config{
+			CompressionAlgo: derive.Zlib,
+		},
 	}, &rollup.Config{})
 	m.Clear(eth.BlockID{})
 
@@ -251,7 +266,9 @@ func TestChannelTxConfirmed(t *testing.T) {
 func TestChannelTxFailed(t *testing.T) {
 	// Create a channel manager
 	log := testlog.Logger(t, log.LevelCrit)
-	m := NewChannelManager(log, metrics.NoopMetrics, ChannelConfig{}, &rollup.Config{})
+	m := NewChannelManager(log, metrics.NoopMetrics, ChannelConfig{CompressorConfig: compressor.Config{
+		CompressionAlgo: derive.Zlib,
+	}}, &rollup.Config{})
 	m.Clear(eth.BlockID{})
 
 	// Let's add a valid pending transaction to the channel

op-batcher/batcher/config.go

@@ -10,6 +10,7 @@ import (
 
 	"github.com/ethereum-optimism/optimism/op-batcher/compressor"
 	"github.com/ethereum-optimism/optimism/op-batcher/flags"
+	"github.com/ethereum-optimism/optimism/op-node/rollup/derive"
 	plasma "github.com/ethereum-optimism/optimism/op-plasma"
 	oplog "github.com/ethereum-optimism/optimism/op-service/log"
 	opmetrics "github.com/ethereum-optimism/optimism/op-service/metrics"
@@ -67,6 +68,9 @@ type CLIConfig struct {
 	// Type of compressor to use. Must be one of [compressor.KindKeys].
 	Compressor string
 
+	// Type of compression algorithm to use. Must be one of [zlib, brotli, brotli[9-11]]
+	CompressionAlgo derive.CompressionAlgo
+
 	// If Stopped is true, the batcher starts stopped and won't start batching right away.
 	// Batching needs to be started via an admin RPC.
 	Stopped bool
@@ -124,6 +128,9 @@ func (c *CLIConfig) Check() error {
 	if c.Compressor == compressor.RatioKind && (c.ApproxComprRatio <= 0 || c.ApproxComprRatio > 1) {
 		return fmt.Errorf("invalid ApproxComprRatio %v for ratio compressor", c.ApproxComprRatio)
 	}
+	if !derive.ValidCompressionAlgoType(c.CompressionAlgo) {
+		return fmt.Errorf("invalid compression algo %v", c.CompressionAlgo)
+	}
 	if c.BatchType > 1 {
 		return fmt.Errorf("unknown batch type: %v", c.BatchType)
 	}
@@ -168,6 +175,7 @@ func NewConfig(ctx *cli.Context) *CLIConfig {
 		TargetNumFrames:              ctx.Int(flags.TargetNumFramesFlag.Name),
 		ApproxComprRatio:             ctx.Float64(flags.ApproxComprRatioFlag.Name),
 		Compressor:                   ctx.String(flags.CompressorFlag.Name),
+		CompressionAlgo:              derive.CompressionAlgo(ctx.String(flags.CompressionAlgoFlag.Name)),
 		Stopped:                      ctx.Bool(flags.StoppedFlag.Name),
 		WaitNodeSync:                 ctx.Bool(flags.WaitNodeSyncFlag.Name),
 		CheckRecentTxsDepth:          ctx.Int(flags.CheckRecentTxsDepthFlag.Name),

op-batcher/batcher/config_test.go

@@ -7,6 +7,7 @@ import (
 	"github.com/ethereum-optimism/optimism/op-batcher/batcher"
 	"github.com/ethereum-optimism/optimism/op-batcher/compressor"
 	"github.com/ethereum-optimism/optimism/op-batcher/flags"
+	"github.com/ethereum-optimism/optimism/op-node/rollup/derive"
 	"github.com/ethereum-optimism/optimism/op-service/log"
 	"github.com/ethereum-optimism/optimism/op-service/metrics"
 	"github.com/ethereum-optimism/optimism/op-service/oppprof"
@@ -36,6 +37,7 @@ func validBatcherConfig() batcher.CLIConfig {
 		PprofConfig:            oppprof.DefaultCLIConfig(),
 		// The compressor config is not checked in config.Check()
 		RPC:             rpc.DefaultCLIConfig(),
+		CompressionAlgo: derive.Zlib,
 	}
 }
 

op-batcher/batcher/service.go

@@ -219,7 +219,7 @@ func (bs *BatcherService) initChannelConfig(cfg *CLIConfig) error {
 		return fmt.Errorf("max frame size %d exceeds plasma max input size %d", cc.MaxFrameSize, plasma.MaxInputSize)
 	}
 
-	cc.InitCompressorConfig(cfg.ApproxComprRatio, cfg.Compressor)
+	cc.InitCompressorConfig(cfg.ApproxComprRatio, cfg.Compressor, cfg.CompressionAlgo)
 
 	if bs.UseBlobs && !bs.RollupConfig.IsEcotone(uint64(time.Now().Unix())) {
 		bs.Log.Error("Cannot use Blob data before Ecotone!") // log only, the batcher may not be actively running.
@@ -228,6 +228,11 @@ func (bs *BatcherService) initChannelConfig(cfg *CLIConfig) error {
 		bs.Log.Warn("Ecotone upgrade is active, but batcher is not configured to use Blobs!")
 	}
 
+	// Checking for brotli compression only post Fjord
+	if bs.ChannelConfig.CompressorConfig.CompressionAlgo.IsBrotli() && !bs.RollupConfig.IsFjord(uint64(time.Now().Unix())) {
+		return fmt.Errorf("cannot use brotli compression before Fjord")
+	}
+
 	if err := cc.Check(); err != nil {
 		return fmt.Errorf("invalid channel configuration: %w", err)
 	}

op-batcher/compressor/config.go

@@ -16,6 +16,9 @@ type Config struct {
 	// Kind of compressor to use. Must be one of KindKeys. If unset, NewCompressor
 	// will default to RatioKind.
 	Kind string
+
+	// Type of compression algorithm to use. Must be one of [zlib, brotli-(9|10|11)]
+	CompressionAlgo derive.CompressionAlgo
 }
 
 func (c Config) NewCompressor() (derive.Compressor, error) {

op-batcher/compressor/ratio_compressor.go

 package compressor
 
 import (
-	"bytes"
-	"compress/zlib"
-
 	"github.com/ethereum-optimism/optimism/op-node/rollup/derive"
 )
 
@@ -11,8 +8,7 @@ type RatioCompressor struct {
 	config Config
 
 	inputBytes int
-	buf        bytes.Buffer
-	compress   *zlib.Writer
+	compressor derive.ChannelCompressor
 }
 
 // NewRatioCompressor creates a new derive.Compressor implementation that uses the target
@@ -25,11 +21,11 @@ func NewRatioCompressor(config Config) (derive.Compressor, error) {
 		config: config,
 	}
 
-	compress, err := zlib.NewWriterLevel(&c.buf, zlib.BestCompression)
+	compressor, err := derive.NewChannelCompressor(config.CompressionAlgo)
 	if err != nil {
 		return nil, err
 	}
-	c.compress = compress
+	c.compressor = compressor
 
 	return c, nil
 }
@@ -39,29 +35,28 @@ func (t *RatioCompressor) Write(p []byte) (int, error) {
 		return 0, err
 	}
 	t.inputBytes += len(p)
-	return t.compress.Write(p)
+	return t.compressor.Write(p)
 }
 
 func (t *RatioCompressor) Close() error {
-	return t.compress.Close()
+	return t.compressor.Close()
 }
 
 func (t *RatioCompressor) Read(p []byte) (int, error) {
-	return t.buf.Read(p)
+	return t.compressor.Read(p)
 }
 
 func (t *RatioCompressor) Reset() {
-	t.buf.Reset()
-	t.compress.Reset(&t.buf)
+	t.compressor.Reset()
 	t.inputBytes = 0
 }
 
 func (t *RatioCompressor) Len() int {
-	return t.buf.Len()
+	return t.compressor.Len()
 }
 
 func (t *RatioCompressor) Flush() error {
-	return t.compress.Flush()
+	return t.compressor.Flush()
 }
 
 func (t *RatioCompressor) FullErr() error {

op-batcher/compressor/ratio_compressor_test.go

@@ -6,6 +6,7 @@ import (
 	"testing"
 
 	"github.com/ethereum-optimism/optimism/op-batcher/compressor"
+	"github.com/ethereum-optimism/optimism/op-node/rollup/derive"
 	"github.com/stretchr/testify/require"
 )
 
@@ -62,6 +63,7 @@ func TestChannelConfig_InputThreshold(t *testing.T) {
 			comp, err := compressor.NewRatioCompressor(compressor.Config{
 				TargetOutputSize: tt.targetOutputSize,
 				ApproxComprRatio: tt.approxComprRatio,
+				CompressionAlgo:  derive.Zlib,
 			})
 			require.NoError(t, err)
 			got := comp.(*compressor.RatioCompressor).InputThreshold()

op-batcher/compressor/shadow_compressor.go

 package compressor
 
 import (
-	"bytes"
-	"compress/zlib"
-
 	"github.com/ethereum-optimism/optimism/op-node/rollup/derive"
 )
 
@@ -21,11 +18,8 @@ const (
 type ShadowCompressor struct {
 	config Config
 
-	buf      bytes.Buffer
-	compress *zlib.Writer
-
-	shadowBuf      bytes.Buffer
-	shadowCompress *zlib.Writer
+	compressor       derive.ChannelCompressor
+	shadowCompressor derive.ChannelCompressor
 
 	fullErr error
 
@@ -45,11 +39,11 @@ func NewShadowCompressor(config Config) (derive.Compressor, error) {
 	}
 
 	var err error
-	c.compress, err = zlib.NewWriterLevel(&c.buf, zlib.BestCompression)
+	c.compressor, err = derive.NewChannelCompressor(config.CompressionAlgo)
 	if err != nil {
 		return nil, err
 	}
-	c.shadowCompress, err = zlib.NewWriterLevel(&c.shadowBuf, zlib.BestCompression)
+	c.shadowCompressor, err = derive.NewChannelCompressor(config.CompressionAlgo)
 	if err != nil {
 		return nil, err
 	}
@@ -62,7 +56,7 @@ func (t *ShadowCompressor) Write(p []byte) (int, error) {
 	if t.fullErr != nil {
 		return 0, t.fullErr
 	}
-	_, err := t.shadowCompress.Write(p)
+	_, err := t.shadowCompressor.Write(p)
 	if err != nil {
 		return 0, err
 	}
@@ -71,10 +65,10 @@ func (t *ShadowCompressor) Write(p []byte) (int, error) {
 		// Do not flush the buffer unless there's some chance we will be over the size limit.
 		// This reduces CPU but more importantly it makes the shadow compression ratio more
 		// closely reflect the ultimate compression ratio.
-		if err = t.shadowCompress.Flush(); err != nil {
+		if err = t.shadowCompressor.Flush(); err != nil {
 			return 0, err
 		}
-		newBound = uint64(t.shadowBuf.Len()) + CloseOverheadZlib
+		newBound = uint64(t.shadowCompressor.Len()) + CloseOverheadZlib
 		if newBound > t.config.TargetOutputSize {
 			t.fullErr = derive.ErrCompressorFull
 			if t.Len() > 0 {
@@ -85,32 +79,30 @@ func (t *ShadowCompressor) Write(p []byte) (int, error) {
 		}
 	}
 	t.bound = newBound
-	return t.compress.Write(p)
+	return t.compressor.Write(p)
 }
 
 func (t *ShadowCompressor) Close() error {
-	return t.compress.Close()
+	return t.compressor.Close()
 }
 
 func (t *ShadowCompressor) Read(p []byte) (int, error) {
-	return t.buf.Read(p)
+	return t.compressor.Read(p)
 }
 
 func (t *ShadowCompressor) Reset() {
-	t.buf.Reset()
-	t.compress.Reset(&t.buf)
-	t.shadowBuf.Reset()
-	t.shadowCompress.Reset(&t.shadowBuf)
+	t.compressor.Reset()
+	t.shadowCompressor.Reset()
 	t.fullErr = nil
 	t.bound = safeCompressionOverhead
 }
 
 func (t *ShadowCompressor) Len() int {
-	return t.buf.Len()
+	return t.compressor.Len()
 }
 
 func (t *ShadowCompressor) Flush() error {
-	return t.compress.Flush()
+	return t.compressor.Flush()
 }
 
 func (t *ShadowCompressor) FullErr() error {

op-batcher/compressor/shadow_compressor_test.go

@@ -63,6 +63,7 @@ func TestShadowCompressor(t *testing.T) {
 
 			sc, err := NewShadowCompressor(Config{
 				TargetOutputSize: test.targetOutputSize,
+				CompressionAlgo:  derive.Zlib,
 			})
 			require.NoError(t, err)
 
@@ -115,6 +116,7 @@ func TestBoundInaccurateForLargeRandomData(t *testing.T) {
 
 	sc, err := NewShadowCompressor(Config{
 		TargetOutputSize: sizeLimit + 100,
+		CompressionAlgo:  derive.Zlib,
 	})
 	require.NoError(t, err)
 

op-batcher/flags/flags.go

@@ -9,6 +9,7 @@ import (
 	"github.com/urfave/cli/v2"
 
 	"github.com/ethereum-optimism/optimism/op-batcher/compressor"
+	"github.com/ethereum-optimism/optimism/op-node/rollup/derive"
 	plasma "github.com/ethereum-optimism/optimism/op-plasma"
 	opservice "github.com/ethereum-optimism/optimism/op-service"
 	openum "github.com/ethereum-optimism/optimism/op-service/enum"
@@ -99,6 +100,15 @@ var (
 			return nil
 		},
 	}
+	CompressionAlgoFlag = &cli.GenericFlag{
+		Name:    "compression-algo",
+		Usage:   "The compression algorithm to use. Valid options: " + openum.EnumString(derive.CompressionAlgoTypes),
+		EnvVars: prefixEnvVars("COMPRESSION_ALGO"),
+		Value: func() *derive.CompressionAlgo {
+			out := derive.Zlib
+			return &out
+		}(),
+	}
 	StoppedFlag = &cli.BoolFlag{
 		Name:    "stopped",
 		Usage:   "Initialize the batcher in a stopped state. The batcher can be started using the admin_startBatcher RPC",
@@ -167,6 +177,7 @@ var optionalFlags = []cli.Flag{
 	BatchTypeFlag,
 	DataAvailabilityTypeFlag,
 	ActiveSequencerCheckDurationFlag,
+	CompressionAlgoFlag,
 }
 
 func init() {

op-e2e/actions/l2_batcher.go

@@ -192,6 +192,7 @@ func (s *L2Batcher) Buffer(t Testing) error {
 			target := batcher.MaxDataSize(1, s.l2BatcherCfg.MaxL1TxSize)
 			c, e := compressor.NewShadowCompressor(compressor.Config{
 				TargetOutputSize: target,
+				CompressionAlgo:  derive.Zlib,
 			})
 			require.NoError(t, e, "failed to create compressor")
 
@@ -200,7 +201,7 @@ func (s *L2Batcher) Buffer(t Testing) error {
 			} else {
 				// use span batch if we're forcing it or if we're at/beyond delta
 				if s.l2BatcherCfg.ForceSubmitSpanBatch || s.rollupCfg.IsDelta(block.Time()) {
-					ch, err = derive.NewSpanChannelOut(s.rollupCfg.Genesis.L2Time, s.rollupCfg.L2ChainID, target)
+					ch, err = derive.NewSpanChannelOut(s.rollupCfg.Genesis.L2Time, s.rollupCfg.L2ChainID, target, derive.Zlib)
 					// use singular batches in all other cases
 				} else {
 					ch, err = derive.NewSingularChannelOut(c)

op-e2e/actions/sync_test.go

@@ -26,7 +26,7 @@ import (
 )
 
 func newSpanChannelOut(t StatefulTesting, e e2eutils.SetupData) derive.ChannelOut {
-	channelOut, err := derive.NewSpanChannelOut(e.RollupCfg.Genesis.L2Time, e.RollupCfg.L2ChainID, 128_000)
+	channelOut, err := derive.NewSpanChannelOut(e.RollupCfg.Genesis.L2Time, e.RollupCfg.L2ChainID, 128_000, derive.Zlib)
 	require.NoError(t, err)
 	return channelOut
 }

op-e2e/sequencer_failover_setup.go

@@ -296,6 +296,7 @@ func setupBatcher(t *testing.T, sys *System, conductors map[string]*conductor) {
 		BatchType:                    derive.SpanBatchType,
 		DataAvailabilityType:         batcherFlags.CalldataType,
 		ActiveSequencerCheckDuration: 0,
+		CompressionAlgo:              derive.Zlib,
 	}
 
 	batcher, err := bss.BatcherServiceFromCLIConfig(context.Background(), "0.0.1", batcherCLIConfig, sys.Cfg.Loggers["batcher"])

op-e2e/setup.go

@@ -845,6 +845,7 @@ func (cfg SystemConfig) Start(t *testing.T, _opts ...SystemConfigOption) (*Syste
 		Stopped:              sys.Cfg.DisableBatcher, // Batch submitter may be enabled later
 		BatchType:            batchType,
 		DataAvailabilityType: sys.Cfg.DataAvailabilityType,
+		CompressionAlgo:      derive.Zlib,
 	}
 	// Batch Submitter
 	batcher, err := bss.BatcherServiceFromCLIConfig(context.Background(), "0.0.1", batcherCLIConfig, sys.Cfg.Loggers["batcher"])

op-node/benchmarks/batchbuilding_test.go

@@ -58,13 +58,13 @@ type compressorAndTarget struct {
 }
 
 // channelOutByType returns a channel out of the given type as a helper for the benchmarks
-func channelOutByType(batchType uint, compKey string) (derive.ChannelOut, error) {
+func channelOutByType(batchType uint, compKey string, algo derive.CompressionAlgo) (derive.ChannelOut, error) {
 	chainID := big.NewInt(333)
 	if batchType == derive.SingularBatchType {
 		return derive.NewSingularChannelOut(compressors[compKey].compressor)
 	}
 	if batchType == derive.SpanBatchType {
-		return derive.NewSpanChannelOut(0, chainID, compressors[compKey].targetOutput)
+		return derive.NewSpanChannelOut(0, chainID, compressors[compKey].targetOutput, algo)
 	}
 	return nil, fmt.Errorf("unsupported batch type: %d", batchType)
 }
@@ -129,13 +129,14 @@ func BenchmarkFinalBatchChannelOut(b *testing.B) {
 			// to leverage optimizations in the Batch Linked List
 			batches[i].Timestamp = uint64(t.Add(time.Duration(i) * time.Second).Unix())
 		}
-		b.Run(tc.String(), func(b *testing.B) {
+		for _, algo := range derive.CompressionAlgoTypes {
+			b.Run(tc.String()+"_"+algo.String(), func(b *testing.B) {
 				// reset the compressor used in the test case
 				for bn := 0; bn < b.N; bn++ {
 					// don't measure the setup time
 					b.StopTimer()
 					compressors[tc.compKey].compressor.Reset()
-				cout, _ := channelOutByType(tc.BatchType, tc.compKey)
+					cout, _ := channelOutByType(tc.BatchType, tc.compKey, algo)
 					// add all but the final batch to the channel out
 					for i := 0; i < tc.BatchCount-1; i++ {
 						err := cout.AddSingularBatch(batches[i], 0)
@@ -149,6 +150,8 @@ func BenchmarkFinalBatchChannelOut(b *testing.B) {
 				}
 			})
 		}
+
+	}
 }
 
 // BenchmarkIncremental fills a channel out incrementally with batches
@@ -165,15 +168,16 @@ func BenchmarkIncremental(b *testing.B) {
 		{derive.SpanBatchType, 5, 1, "RealBlindCompressor"},
 		//{derive.SingularBatchType, 100, 1, "RealShadowCompressor"},
 	}
+	for _, algo := range derive.CompressionAlgoTypes {
 		for _, tc := range tcs {
-		cout, err := channelOutByType(tc.BatchType, tc.compKey)
+			cout, err := channelOutByType(tc.BatchType, tc.compKey, algo)
 			if err != nil {
 				b.Fatal(err)
 			}
 			done := false
 			for base := 0; !done; base += tc.BatchCount {
-			rangeName := fmt.Sprintf("Incremental %s: %d-%d", tc.String(), base, base+tc.BatchCount)
-			b.Run(rangeName, func(b *testing.B) {
+				rangeName := fmt.Sprintf("Incremental %s-%s: %d-%d", algo, tc.String(), base, base+tc.BatchCount)
+				b.Run(rangeName+"_"+algo.String(), func(b *testing.B) {
 					b.StopTimer()
 					// prepare the batches
 					t := time.Now()
@@ -196,6 +200,7 @@ func BenchmarkIncremental(b *testing.B) {
 				})
 			}
 		}
+	}
 }
 
 // BenchmarkAllBatchesChannelOut benchmarks the performance of adding singular batches to a channel out
@@ -226,6 +231,7 @@ func BenchmarkAllBatchesChannelOut(b *testing.B) {
 		}
 	}
 
+	for _, algo := range derive.CompressionAlgoTypes {
 		for _, tc := range tests {
 			chainID := big.NewInt(333)
 			rng := rand.New(rand.NewSource(0x543331))
@@ -238,13 +244,13 @@ func BenchmarkAllBatchesChannelOut(b *testing.B) {
 				// to leverage optimizations in the Batch Linked List
 				batches[i].Timestamp = uint64(t.Add(time.Duration(i) * time.Second).Unix())
 			}
-		b.Run(tc.String(), func(b *testing.B) {
+			b.Run(tc.String()+"_"+algo.String(), func(b *testing.B) {
 				// reset the compressor used in the test case
 				for bn := 0; bn < b.N; bn++ {
 					// don't measure the setup time
 					b.StopTimer()
 					compressors[tc.compKey].compressor.Reset()
-				cout, _ := channelOutByType(tc.BatchType, tc.compKey)
+					cout, _ := channelOutByType(tc.BatchType, tc.compKey, algo)
 					b.StartTimer()
 					// add all batches to the channel out
 					for i := 0; i < tc.BatchCount; i++ {
@@ -254,6 +260,7 @@ func BenchmarkAllBatchesChannelOut(b *testing.B) {
 				}
 			})
 		}
+	}
 }
 
 // BenchmarkGetRawSpanBatch benchmarks the performance of building a span batch from singular batches

op-node/cmd/batch_decoder/reassemble/reassemble.go

@@ -111,7 +111,7 @@ func processFrames(cfg Config, rollupCfg *rollup.Config, id derive.ChannelID, fr
 	var batchTypes []int
 	invalidBatches := false
 	if ch.IsReady() {
-		br, err := derive.BatchReader(ch.Reader(), spec.MaxRLPBytesPerChannel(ch.HighestBlock().Time))
+		br, err := derive.BatchReader(ch.Reader(), spec.MaxRLPBytesPerChannel(ch.HighestBlock().Time), rollupCfg.IsFjord(ch.HighestBlock().Time))
 		if err == nil {
 			for batchData, err := br(); err != io.EOF; batchData, err = br() {
 				if err != nil {

op-node/rollup/derive/channel.go

 package derive
 
 import (
+	"bufio"
 	"bytes"
 	"compress/zlib"
 	"fmt"
 	"io"
 
+	"github.com/andybalholm/brotli"
 	"github.com/ethereum-optimism/optimism/op-service/eth"
 	"github.com/ethereum/go-ethereum/rlp"
 )
 
+const (
+	ZlibCM8  = 8
+	ZlibCM15 = 15
+)
+
 // A Channel is a set of batches that are split into at least one, but possibly multiple frames.
 // Frames are allowed to be ingested out of order.
 // Each frame is ingested one by one. Once a frame with `closed` is added to the channel, the
@@ -151,17 +158,44 @@ func (ch *Channel) Reader() io.Reader {
 // The L1Inclusion block is also provided at creation time.
 // Warning: the batch reader can read every batch-type.
 // The caller of the batch-reader should filter the results.
-func BatchReader(r io.Reader, maxRLPBytesPerChannel uint64) (func() (*BatchData, error), error) {
-	// Setup decompressor stage + RLP reader
-	zr, err := zlib.NewReader(r)
+func BatchReader(r io.Reader, maxRLPBytesPerChannel uint64, isFjord bool) (func() (*BatchData, error), error) {
+	// use buffered reader so can peek the first byte
+	bufReader := bufio.NewReader(r)
+	compressionType, err := bufReader.Peek(1)
+	if err != nil {
+		return nil, err
+	}
+
+	var zr io.Reader
+	// For zlib, the last 4 bits must be either 8 or 15 (both are reserved value)
+	if compressionType[0]&0x0F == ZlibCM8 || compressionType[0]&0x0F == ZlibCM15 {
+		var err error
+		zr, err = zlib.NewReader(bufReader)
 		if err != nil {
 			return nil, err
 		}
+		// If the bits equal to 1, then it is a brotli reader
+	} else if compressionType[0] == ChannelVersionBrotli {
+		// If before Fjord, we cannot accept brotli compressed batch
+		if !isFjord {
+			return nil, fmt.Errorf("cannot accept brotli compressed batch before Fjord")
+		}
+		// discard the first byte
+		_, err := bufReader.Discard(1)
+		if err != nil {
+			return nil, err
+		}
+		zr = brotli.NewReader(bufReader)
+	} else {
+		return nil, fmt.Errorf("cannot distinguish the compression algo used given type byte %v", compressionType[0])
+	}
+
+	// Setup decompressor stage + RLP reader
 	rlpReader := rlp.NewStream(zr, maxRLPBytesPerChannel)
 	// Read each batch iteratively
 	return func() (*BatchData, error) {
 		var batchData BatchData
-		if err = rlpReader.Decode(&batchData); err != nil {
+		if err := rlpReader.Decode(&batchData); err != nil {
 			return nil, err
 		}
 		return &batchData, nil

op-node/rollup/derive/channel_compressor.go

+package derive
+
+import (
+	"bytes"
+	"compress/zlib"
+	"fmt"
+	"io"
+
+	"github.com/andybalholm/brotli"
+)
+
+const (
+	ChannelVersionBrotli byte = 0x01
+)
+
+type ChannelCompressor interface {
+	Write([]byte) (int, error)
+	Flush() error
+	Close() error
+	Reset()
+	Len() int
+	Read([]byte) (int, error)
+	GetCompressed() *bytes.Buffer
+}
+
+type CompressorWriter interface {
+	Write([]byte) (int, error)
+	Flush() error
+	Close() error
+	Reset(io.Writer)
+}
+
+type BaseChannelCompressor struct {
+	compressed *bytes.Buffer
+	CompressorWriter
+}
+
+func (bcc *BaseChannelCompressor) Len() int {
+	return bcc.compressed.Len()
+}
+
+func (bcc *BaseChannelCompressor) Read(p []byte) (int, error) {
+	return bcc.compressed.Read(p)
+}
+
+func (bcc *BaseChannelCompressor) GetCompressed() *bytes.Buffer {
+	return bcc.compressed
+}
+
+type ZlibCompressor struct {
+	BaseChannelCompressor
+}
+
+func (zc *ZlibCompressor) Reset() {
+	zc.compressed.Reset()
+	zc.CompressorWriter.Reset(zc.compressed)
+}
+
+type BrotliCompressor struct {
+	BaseChannelCompressor
+}
+
+func (bc *BrotliCompressor) Reset() {
+	bc.compressed.Reset()
+	bc.compressed.WriteByte(ChannelVersionBrotli)
+	bc.CompressorWriter.Reset(bc.compressed)
+}
+
+func NewChannelCompressor(algo CompressionAlgo) (ChannelCompressor, error) {
+	compressed := &bytes.Buffer{}
+	if algo == Zlib {
+		writer, err := zlib.NewWriterLevel(compressed, zlib.BestCompression)
+		if err != nil {
+			return nil, err
+		}
+		return &ZlibCompressor{
+			BaseChannelCompressor{
+				CompressorWriter: writer,
+				compressed:       compressed,
+			},
+		}, nil
+	} else if algo.IsBrotli() {
+		compressed.WriteByte(ChannelVersionBrotli)
+		writer := brotli.NewWriterLevel(compressed, GetBrotliLevel(algo))
+		return &BrotliCompressor{
+			BaseChannelCompressor{
+				CompressorWriter: writer,
+				compressed:       compressed,
+			},
+		}, nil
+	} else {
+		return nil, fmt.Errorf("unsupported compression algorithm: %s", algo)
+	}
+}

op-node/rollup/derive/channel_compressor_test.go

+package derive
+
+import (
+	"math/rand"
+	"testing"
+
+	"github.com/stretchr/testify/require"
+)
+
+var r = rand.New(rand.NewSource(99))
+
+func randomBytes(length int) []byte {
+	b := make([]byte, length)
+	_, err := r.Read(b)
+	// Rand.Read always returns nil error
+	if err != nil {
+		panic(err)
+	}
+	return b
+}
+
+func TestChannelCompressor_NewReset(t *testing.T) {
+	testCases := []struct {
+		name              string
+		algo              CompressionAlgo
+		expectedResetSize int
+		expectErr         bool
+	}{
+		{
+			name:              "zlib",
+			algo:              Zlib,
+			expectedResetSize: 0,
+		},
+		{
+			name:              "brotli10",
+			algo:              Brotli10,
+			expectedResetSize: 1,
+		},
+		{
+			name:              "zstd",
+			algo:              CompressionAlgo("zstd"),
+			expectedResetSize: 0,
+			expectErr:         true,
+		},
+	}
+
+	for _, tc := range testCases {
+		t.Run(tc.name, func(t *testing.T) {
+			scc, err := NewChannelCompressor(tc.algo)
+			if tc.expectErr {
+				require.Error(t, err)
+				return
+			}
+			require.NoError(t, err)
+			require.Equal(t, tc.expectedResetSize, scc.Len())
+
+			_, err = scc.Write(randomBytes(10))
+			require.NoError(t, err)
+			err = scc.Flush()
+			require.NoError(t, err)
+			require.Greater(t, scc.Len(), tc.expectedResetSize)
+
+			scc.Reset()
+			require.Equal(t, tc.expectedResetSize, scc.Len())
+		})
+	}
+}

op-node/rollup/derive/channel_in_reader.go

@@ -44,7 +44,7 @@ func (cr *ChannelInReader) Origin() eth.L1BlockRef {
 
 // TODO: Take full channel for better logging
 func (cr *ChannelInReader) WriteChannel(data []byte) error {
-	if f, err := BatchReader(bytes.NewBuffer(data), cr.spec.MaxRLPBytesPerChannel(cr.prev.Origin().Time)); err == nil {
+	if f, err := BatchReader(bytes.NewBuffer(data), cr.spec.MaxRLPBytesPerChannel(cr.prev.Origin().Time), cr.cfg.IsFjord(cr.prev.Origin().Time)); err == nil {
 		cr.nextBatchFn = f
 		cr.metrics.RecordChannelInputBytes(len(data))
 		return nil

op-node/rollup/derive/channel_out_test.go

@@ -52,7 +52,7 @@ var channelTypes = []struct {
 	{
 		Name: "Span",
 		ChannelOut: func(t *testing.T) ChannelOut {
-			cout, err := NewSpanChannelOut(0, big.NewInt(0), 128_000)
+			cout, err := NewSpanChannelOut(0, big.NewInt(0), 128_000, Zlib)
 			require.NoError(t, err)
 			return cout
 		},
@@ -113,7 +113,7 @@ func TestOutputFrameNoEmptyLastFrame(t *testing.T) {
 
 			// depending on the channel type, determine the size of the written data
 			if span, ok := cout.(*SpanChannelOut); ok {
-				written = uint64(span.compressed.Len())
+				written = uint64(span.compressor.Len())
 			} else if singular, ok := cout.(*SingularChannelOut); ok {
 				written = uint64(singular.compress.Len())
 			}
@@ -220,12 +220,12 @@ func TestBlockToBatchValidity(t *testing.T) {
 	require.ErrorContains(t, err, "has no transactions")
 }
 
-func SpanChannelAndBatches(t *testing.T, target uint64, len int) (*SpanChannelOut, []*SingularBatch) {
+func SpanChannelAndBatches(t *testing.T, target uint64, len int, algo CompressionAlgo) (*SpanChannelOut, []*SingularBatch) {
 	// target is larger than one batch, but smaller than two batches
 	rng := rand.New(rand.NewSource(0x543331))
 	chainID := big.NewInt(rng.Int63n(1000))
 	txCount := 1
-	cout, err := NewSpanChannelOut(0, chainID, target)
+	cout, err := NewSpanChannelOut(0, chainID, target, algo)
 	require.NoError(t, err)
 	batches := make([]*SingularBatch, len)
 	// adding the first batch should not cause an error
@@ -237,14 +237,33 @@ func SpanChannelAndBatches(t *testing.T, target uint64, len int) (*SpanChannelOu
 	return cout, batches
 }
 
+func TestSpanChannelOut(t *testing.T) {
+	tests := []struct {
+		name string
+		f    func(t *testing.T, algo CompressionAlgo)
+	}{
+		{"SpanChannelOutCompressionOnlyOneBatch", SpanChannelOutCompressionOnlyOneBatch},
+		{"SpanChannelOutCompressionUndo", SpanChannelOutCompressionUndo},
+		{"SpanChannelOutClose", SpanChannelOutClose},
+	}
+	for _, test := range tests {
+		test := test
+		for _, algo := range CompressionAlgoTypes {
+			t.Run(test.name+"_"+algo.String(), func(t *testing.T) {
+				test.f(t, algo)
+			})
+		}
+	}
+}
+
 // TestSpanChannelOutCompressionOnlyOneBatch tests that the SpanChannelOut compression works as expected when there is only one batch
 // and it is larger than the target size. The single batch should be compressed, and the channel should now be full
-func TestSpanChannelOutCompressionOnlyOneBatch(t *testing.T) {
-	cout, singularBatches := SpanChannelAndBatches(t, 300, 2)
+func SpanChannelOutCompressionOnlyOneBatch(t *testing.T, algo CompressionAlgo) {
+	cout, singularBatches := SpanChannelAndBatches(t, 300, 2, algo)
 
 	err := cout.AddSingularBatch(singularBatches[0], 0)
 	// confirm compression was not skipped
-	require.Greater(t, cout.compressed.Len(), 0)
+	require.Greater(t, cout.compressor.Len(), 0)
 	require.NoError(t, err)
 
 	// confirm the channel is full
@@ -256,21 +275,25 @@ func TestSpanChannelOutCompressionOnlyOneBatch(t *testing.T) {
 }
 
 // TestSpanChannelOutCompressionUndo tests that the SpanChannelOut compression rejects a batch that would cause the channel to be overfull
-func TestSpanChannelOutCompressionUndo(t *testing.T) {
+func SpanChannelOutCompressionUndo(t *testing.T, algo CompressionAlgo) {
 	// target is larger than one batch, but smaller than two batches
-	cout, singularBatches := SpanChannelAndBatches(t, 750, 2)
+	cout, singularBatches := SpanChannelAndBatches(t, 750, 2, algo)
 
 	err := cout.AddSingularBatch(singularBatches[0], 0)
 	require.NoError(t, err)
 	// confirm that the first compression was skipped
-	require.Equal(t, 0, cout.compressed.Len())
+	if algo == Zlib {
+		require.Equal(t, 0, cout.compressor.Len())
+	} else {
+		require.Equal(t, 1, cout.compressor.Len()) // 1 because of brotli channel version
+	}
 	// record the RLP length to confirm it doesn't change when adding a rejected batch
 	rlp1 := cout.activeRLP().Len()
 
 	err = cout.AddSingularBatch(singularBatches[1], 0)
 	require.ErrorIs(t, err, ErrCompressorFull)
 	// confirm that the second compression was not skipped
-	require.Greater(t, cout.compressed.Len(), 0)
+	require.Greater(t, cout.compressor.Len(), 0)
 
 	// confirm that the second rlp is tht same size as the first (because the second batch was not added)
 	require.Equal(t, rlp1, cout.activeRLP().Len())
@@ -278,14 +301,19 @@ func TestSpanChannelOutCompressionUndo(t *testing.T) {
 
 // TestSpanChannelOutClose tests that the SpanChannelOut compression works as expected when the channel is closed.
 // it should compress the batch even if it is smaller than the target size because the channel is closing
-func TestSpanChannelOutClose(t *testing.T) {
+func SpanChannelOutClose(t *testing.T, algo CompressionAlgo) {
 	target := uint64(600)
-	cout, singularBatches := SpanChannelAndBatches(t, target, 1)
+	cout, singularBatches := SpanChannelAndBatches(t, target, 1, algo)
 
 	err := cout.AddSingularBatch(singularBatches[0], 0)
 	require.NoError(t, err)
 	// confirm no compression has happened yet
-	require.Equal(t, 0, cout.compressed.Len())
+
+	if algo == Zlib {
+		require.Equal(t, 0, cout.compressor.Len())
+	} else {
+		require.Equal(t, 1, cout.compressor.Len()) // 1 because of brotli channel version
+	}
 
 	// confirm the RLP length is less than the target
 	rlpLen := cout.activeRLP().Len()
@@ -295,6 +323,6 @@ func TestSpanChannelOutClose(t *testing.T) {
 	require.NoError(t, cout.Close())
 
 	// confirm that the only batch was compressed, and that the RLP did not change
-	require.Greater(t, cout.compressed.Len(), 0)
+	require.Greater(t, cout.compressor.Len(), 0)
 	require.Equal(t, rlpLen, cout.activeRLP().Len())
 }

op-node/rollup/derive/channel_test.go

 package derive
 
 import (
+	"bytes"
+	"compress/zlib"
+	"math/big"
+	"math/rand"
 	"testing"
 
+	"github.com/DataDog/zstd"
+	"github.com/andybalholm/brotli"
 	"github.com/ethereum-optimism/optimism/op-service/eth"
 	"github.com/stretchr/testify/require"
 )
@@ -99,3 +105,117 @@ func TestFrameValidity(t *testing.T) {
 		t.Run(tc.name, tc.Run)
 	}
 }
+
+func TestBatchReader(t *testing.T) {
+	// Get batch data
+	rng := rand.New(rand.NewSource(0x543331))
+	singularBatch := RandomSingularBatch(rng, 20, big.NewInt(333))
+	batchDataInput := NewBatchData(singularBatch)
+
+	encodedBatch := &bytes.Buffer{}
+	err := batchDataInput.EncodeRLP(encodedBatch)
+	require.NoError(t, err)
+
+	var testCases = []struct {
+		name      string
+		algo      func(buf *bytes.Buffer, t *testing.T)
+		isFjord   bool
+		expectErr bool
+	}{
+		{
+			name: "zlib-post-fjord",
+			algo: func(buf *bytes.Buffer, t *testing.T) {
+				writer := zlib.NewWriter(buf)
+				_, err := writer.Write(encodedBatch.Bytes())
+				require.NoError(t, err)
+				writer.Close()
+			},
+			isFjord: true,
+		},
+		{
+			name: "zlib-pre-fjord",
+			algo: func(buf *bytes.Buffer, t *testing.T) {
+				writer := zlib.NewWriter(buf)
+				_, err := writer.Write(encodedBatch.Bytes())
+				require.NoError(t, err)
+				writer.Close()
+			},
+			isFjord: false,
+		},
+		{
+			name: "brotli9-post-fjord",
+			algo: func(buf *bytes.Buffer, t *testing.T) {
+				buf.WriteByte(ChannelVersionBrotli)
+				writer := brotli.NewWriterLevel(buf, 9)
+				_, err := writer.Write(encodedBatch.Bytes())
+				require.NoError(t, err)
+				writer.Close()
+			},
+			isFjord: true,
+		},
+		{
+			name: "brotli9-pre-fjord",
+			algo: func(buf *bytes.Buffer, t *testing.T) {
+				buf.WriteByte(ChannelVersionBrotli)
+				writer := brotli.NewWriterLevel(buf, 9)
+				_, err := writer.Write(encodedBatch.Bytes())
+				require.NoError(t, err)
+				writer.Close()
+			},
+			isFjord:   false,
+			expectErr: true, // expect an error because brotli is not supported before Fjord
+		},
+		{
+			name: "brotli10-post-fjord",
+			algo: func(buf *bytes.Buffer, t *testing.T) {
+				buf.WriteByte(ChannelVersionBrotli)
+				writer := brotli.NewWriterLevel(buf, 10)
+				_, err := writer.Write(encodedBatch.Bytes())
+				require.NoError(t, err)
+				writer.Close()
+			},
+			isFjord: true,
+		},
+		{
+			name: "brotli11-post-fjord",
+			algo: func(buf *bytes.Buffer, t *testing.T) {
+				buf.WriteByte(ChannelVersionBrotli)
+				writer := brotli.NewWriterLevel(buf, 11)
+				_, err := writer.Write(encodedBatch.Bytes())
+				require.NoError(t, err)
+				writer.Close()
+			},
+			isFjord: true,
+		},
+		{
+			name: "zstd-post-fjord",
+			algo: func(buf *bytes.Buffer, t *testing.T) {
+				writer := zstd.NewWriter(buf)
+				_, err := writer.Write(encodedBatch.Bytes())
+				require.NoError(t, err)
+				writer.Close()
+			},
+			expectErr: true,
+			isFjord:   true,
+		}}
+
+	for _, tc := range testCases {
+		compressed := new(bytes.Buffer)
+		tc := tc
+		t.Run(tc.name, func(t *testing.T) {
+			tc.algo(compressed, t)
+			reader, err := BatchReader(bytes.NewReader(compressed.Bytes()), 120000, tc.isFjord)
+			if tc.expectErr {
+				require.Error(t, err)
+				return
+			}
+			require.NoError(t, err)
+
+			// read the batch data
+			batchData, err := reader()
+			require.NoError(t, err)
+			require.NotNil(t, batchData)
+			require.Equal(t, batchDataInput, batchData)
+		})
+	}
+}

op-node/rollup/derive/span_channel_out.go

@@ -2,7 +2,7 @@ package derive
 
 import (
 	"bytes"
-	"compress/zlib"
+
 	"crypto/rand"
 	"fmt"
 	"io"
@@ -26,10 +26,8 @@ type SpanChannelOut struct {
 	// lastCompressedRLPSize tracks the *uncompressed* size of the last RLP buffer that was compressed
 	// it is used to measure the growth of the RLP buffer when adding a new batch to optimize compression
 	lastCompressedRLPSize int
-	// compressed contains compressed data for making output frames
-	compressed *bytes.Buffer
-	// compress is the zlib writer for the channel
-	compressor *zlib.Writer
+	// the compressor for the channel
+	compressor ChannelCompressor
 	// target is the target size of the compressed data
 	target uint64
 	// closed indicates if the channel is closed
@@ -49,22 +47,23 @@ func (co *SpanChannelOut) setRandomID() error {
 	return err
 }
 
-func NewSpanChannelOut(genesisTimestamp uint64, chainID *big.Int, targetOutputSize uint64) (*SpanChannelOut, error) {
+func NewSpanChannelOut(genesisTimestamp uint64, chainID *big.Int, targetOutputSize uint64, compressionAlgo CompressionAlgo) (*SpanChannelOut, error) {
 	c := &SpanChannelOut{
 		id:        ChannelID{},
 		frame:     0,
 		spanBatch: NewSpanBatch(genesisTimestamp, chainID),
 		rlp:       [2]*bytes.Buffer{{}, {}},
-		compressed: &bytes.Buffer{},
 		target:    targetOutputSize,
 	}
 	var err error
 	if err = c.setRandomID(); err != nil {
 		return nil, err
 	}
-	if c.compressor, err = zlib.NewWriterLevel(c.compressed, zlib.BestCompression); err != nil {
+
+	if c.compressor, err = NewChannelCompressor(compressionAlgo); err != nil {
 		return nil, err
 	}
+
 	return c, nil
 }
 
@@ -75,8 +74,7 @@ func (co *SpanChannelOut) Reset() error {
 	co.rlp[0].Reset()
 	co.rlp[1].Reset()
 	co.lastCompressedRLPSize = 0
-	co.compressed.Reset()
-	co.compressor.Reset(co.compressed)
+	co.compressor.Reset()
 	co.spanBatch = NewSpanBatch(co.spanBatch.GenesisTimestamp, co.spanBatch.ChainID)
 	// setting the new randomID is the only part of the reset that can fail
 	return co.setRandomID()
@@ -153,7 +151,7 @@ func (co *SpanChannelOut) AddSingularBatch(batch *SingularBatch, seqNum uint64)
 	// if the compressed data *plus* the new rlp data is under the target size, return early
 	// this optimizes out cases where the compressor will obviously come in under the target size
 	rlpGrowth := co.activeRLP().Len() - co.lastCompressedRLPSize
-	if uint64(co.compressed.Len()+rlpGrowth) < co.target {
+	if uint64(co.compressor.Len()+rlpGrowth) < co.target {
 		return nil
 	}
 
@@ -186,8 +184,7 @@ func (co *SpanChannelOut) AddSingularBatch(batch *SingularBatch, seqNum uint64)
 // compress compresses the active RLP buffer and checks if the compressed data is over the target size.
 // it resets all the compression buffers because Span Batches aren't meant to be compressed incrementally.
 func (co *SpanChannelOut) compress() error {
-	co.compressed.Reset()
-	co.compressor.Reset(co.compressed)
+	co.compressor.Reset()
 	if _, err := co.compressor.Write(co.activeRLP().Bytes()); err != nil {
 		return err
 	}
@@ -207,7 +204,7 @@ func (co *SpanChannelOut) InputBytes() int {
 // Span Channel Out does not provide early output, so this will always be 0 until the channel is closed or full
 func (co *SpanChannelOut) ReadyBytes() int {
 	if co.closed || co.FullErr() != nil {
-		return co.compressed.Len()
+		return co.compressor.Len()
 	}
 	return 0
 }
@@ -225,7 +222,7 @@ func (co *SpanChannelOut) checkFull() {
 	if co.full != nil {
 		return
 	}
-	if uint64(co.compressed.Len()) >= co.target {
+	if uint64(co.compressor.Len()) >= co.target {
 		co.full = ErrCompressorFull
 	}
 }
@@ -264,7 +261,7 @@ func (co *SpanChannelOut) OutputFrame(w *bytes.Buffer, maxSize uint64) (uint16,
 
 	f := createEmptyFrame(co.id, co.frame, co.ReadyBytes(), co.closed, maxSize)
 
-	if _, err := io.ReadFull(co.compressed, f.Data); err != nil {
+	if _, err := io.ReadFull(co.compressor.GetCompressed(), f.Data); err != nil {
 		return 0, err
 	}
 

op-node/rollup/derive/types.go

+package derive
+
+import (
+	"fmt"
+	"regexp"
+)
+
+type CompressionAlgo string
+
+const (
+	// compression algo types
+	Zlib     CompressionAlgo = "zlib"
+	Brotli9  CompressionAlgo = "brotli-9"
+	Brotli10 CompressionAlgo = "brotli-10"
+	Brotli11 CompressionAlgo = "brotli-11"
+)
+
+var CompressionAlgoTypes = []CompressionAlgo{
+	Zlib,
+	Brotli9,
+	Brotli10,
+	Brotli11,
+}
+
+var brotliRegexp = regexp.MustCompile(`^brotli-(9|10|11)$`)
+
+func (algo CompressionAlgo) String() string {
+	return string(algo)
+}
+
+func (algo *CompressionAlgo) Set(value string) error {
+	if !ValidCompressionAlgoType(CompressionAlgo(value)) {
+		return fmt.Errorf("unknown compression algo type: %q", value)
+	}
+	*algo = CompressionAlgo(value)
+	return nil
+}
+
+func (algo *CompressionAlgo) Clone() any {
+	cpy := *algo
+	return &cpy
+}
+
+func (algo *CompressionAlgo) IsBrotli() bool {
+	return brotliRegexp.MatchString(algo.String())
+}
+
+func GetBrotliLevel(algo CompressionAlgo) int {
+	switch algo {
+	case Brotli9:
+		return 9
+	case Brotli10:
+		return 10
+	case Brotli11:
+		return 11
+	default:
+		panic("Unsupported brotli level")
+	}
+}
+
+func ValidCompressionAlgoType(value CompressionAlgo) bool {
+	for _, k := range CompressionAlgoTypes {
+		if k == value {
+			return true
+		}
+	}
+	return false
+}

op-node/rollup/derive/types_test.go

+package derive
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/require"
+)
+
+func TestCompressionAlgo(t *testing.T) {
+	testCases := []struct {
+		name                       string
+		algo                       CompressionAlgo
+		isBrotli                   bool
+		isValidCompressionAlgoType bool
+	}{
+		{
+			name:                       "zlib",
+			algo:                       Zlib,
+			isBrotli:                   false,
+			isValidCompressionAlgoType: true,
+		},
+		{
+			name:                       "brotli-9",
+			algo:                       Brotli9,
+			isBrotli:                   true,
+			isValidCompressionAlgoType: true,
+		},
+		{
+			name:                       "brotli-10",
+			algo:                       Brotli10,
+			isBrotli:                   true,
+			isValidCompressionAlgoType: true,
+		},
+		{
+			name:                       "brotli-11",
+			algo:                       Brotli11,
+			isBrotli:                   true,
+			isValidCompressionAlgoType: true,
+		},
+		{
+			name:                       "invalid",
+			algo:                       CompressionAlgo("invalid"),
+			isBrotli:                   false,
+			isValidCompressionAlgoType: false,
+		}}
+
+	for _, tc := range testCases {
+		t.Run(tc.name, func(t *testing.T) {
+			require.Equal(t, tc.isBrotli, tc.algo.IsBrotli())
+			if tc.isBrotli {
+				require.NotPanics(t, func() { GetBrotliLevel((tc.algo)) })
+			} else {
+				require.Panics(t, func() { GetBrotliLevel(tc.algo) })
+			}
+			require.Equal(t, tc.isValidCompressionAlgoType, ValidCompressionAlgoType(tc.algo))
+		})
+	}
+}