[storage][azblob]DownloadFile: Download large files serially

sdk/storage/azblob/blob/client.go

@@ -8,12 +8,13 @@ package blob
 
 import (
 	"context"
-	"github.com/Azure/azure-sdk-for-go/sdk/storage/azblob/bloberror"
 	"io"
 	"os"
 	"sync"
 	"time"
 
+	"github.com/Azure/azure-sdk-for-go/sdk/storage/azblob/bloberror"
+
 	"github.com/Azure/azure-sdk-for-go/sdk/azcore"
 	"github.com/Azure/azure-sdk-for-go/sdk/azcore/runtime"
 	"github.com/Azure/azure-sdk-for-go/sdk/azcore/streaming"
@@ -314,8 +315,8 @@ func (b *Client) GetSASURL(permissions sas.BlobPermissions, expiry time.Time, o
 
 // Concurrent Download Functions -----------------------------------------------------------------------------------------
 
-// download downloads an Azure blob to a WriterAt in parallel.
-func (b *Client) download(ctx context.Context, writer io.WriterAt, o downloadOptions) (int64, error) {
+// downloadBuffer downloads an Azure blob to a WriterAt in parallel.
+func (b *Client) downloadBuffer(ctx context.Context, writer io.WriterAt, o downloadOptions) (int64, error) {
 	if o.BlockSize == 0 {
 		o.BlockSize = DefaultDownloadBlockSize
 	}
@@ -381,6 +382,157 @@ func (b *Client) download(ctx context.Context, writer io.WriterAt, o downloadOpt
 	return count, nil
 }
 
+// downloadFile downloads an Azure blob to a Writer. The blocks are downloaded parallely,
+// but written to file serially
+func (b *Client) downloadFile(ctx context.Context, writer io.Writer, o downloadOptions) (int64, error) {
+	ctx, cancel := context.WithCancel(ctx)
+	defer cancel()
+	if o.BlockSize == 0 {
+		o.BlockSize = DefaultDownloadBlockSize
+	}
+
+	if o.Concurrency == 0 {
+		o.Concurrency = defaultConcurrency
+	}
+
+	count := o.Range.Count
+	if count == CountToEnd { // If size not specified, calculate it
+		// If we don't have the length at all, get it
+		gr, err := b.GetProperties(ctx, o.getBlobPropertiesOptions())
+		if err != nil {
+			return 0, err
+		}
+		count = *gr.ContentLength - o.Range.Offset
+	}
+
+	if count <= 0 {
+		// The file is empty, there is nothing to download.
+		return 0, nil
+	}
+
+	progress := int64(0)
+	progressLock := &sync.Mutex{}
+
+	// helper routine to get body
+	getBodyForRange := func(ctx context.Context, chunkStart, size int64) (io.ReadCloser, error) {
+		downloadBlobOptions := o.getDownloadBlobOptions(HTTPRange{
+			Offset: chunkStart + o.Range.Offset,
+			Count:  size,
+		}, nil)
+		dr, err := b.DownloadStream(ctx, downloadBlobOptions)
+		if err != nil {
+			return nil, err
+		}
+
+		var body io.ReadCloser = dr.NewRetryReader(ctx, &o.RetryReaderOptionsPerBlock)
+		if o.Progress != nil {
+			rangeProgress := int64(0)
+			body = streaming.NewResponseProgress(
+				body,
+				func(bytesTransferred int64) {
+					diff := bytesTransferred - rangeProgress
+					rangeProgress = bytesTransferred
+					progressLock.Lock()
+					progress += diff
+					o.Progress(progress)
+					progressLock.Unlock()
+				})
+		}
+
+		return body, nil
+	}
+
+	// if file fits in a single buffer, we'll download here.
+	if count <= o.BlockSize {
+		body, err := getBodyForRange(ctx, int64(0), count)
+		if err != nil {
+			return 0, err
+		}
+
+		return io.Copy(writer, body)
+	}
+
+	buffers := shared.NewMMBPool(int(o.Concurrency), o.BlockSize)
+	defer buffers.Free()
+	aquireBuffer := func() ([]byte, error) {
+		select {
+		case b := <-buffers.Acquire():
+			// got a buffer
+			return b, nil
+		default:
+			// no buffer available; allocate a new buffer if possible
+			if _, err := buffers.Grow(); err != nil {
+				return nil, err
+			}
+
+			// either grab the newly allocated buffer or wait for one to become available
+			return <-buffers.Acquire(), nil
+		}
+	}
+
+	numChunks := uint16((count-1)/o.BlockSize) + 1
+	blocks := make([]chan []byte, numChunks)
+	for b := range blocks {
+		blocks[b] = make(chan []byte)
+	}
+
+	writerError := make(chan error)
+	go func(ch chan error) {
+		for _, block := range blocks {
+			select {
+			case <-ctx.Done():
+				return
+			case block := <-block:
+				_, err := writer.Write(block)
+				buffers.Release(block)
+				if err != nil {
+					ch <- err
+					return
+				}
+			}
+		}
+		ch <- nil
+	}(writerError)
+
+	// Prepare and do parallel download.
+	err := shared.DoBatchTransfer(ctx, &shared.BatchTransferOptions{
+		OperationName: "downloadBlobToWriterAt",
+		TransferSize:  count,
+		ChunkSize:     o.BlockSize,
+		Concurrency:   o.Concurrency,
+		Operation: func(ctx context.Context, chunkStart int64, count int64) error {
+			body, err := getBodyForRange(ctx, chunkStart, count)
+			if err != nil {
+				return nil
+			}
+
+			buff, err := aquireBuffer()
+			if err != nil {
+				return err
+			}
+
+			_, err = io.ReadFull(body, buff[:count])
+			if err != nil {
+				return err
+			}
+			body.Close()
+
+			blockIndex := (chunkStart / o.BlockSize)
+			blocks[blockIndex] <- buff
+			return nil
+		},
+	})
+
+	if err != nil {
+		return 0, err
+	}
+	// error from writer thread.
+	if err = <-writerError; err != nil {
+		return 0, err
+	}
+	return count, nil
+}
+
 // DownloadStream reads a range of bytes from a blob. The response also includes the blob's properties and metadata.
 // For more information, see https://docs.microsoft.com/rest/api/storageservices/get-blob.
 func (b *Client) DownloadStream(ctx context.Context, o *DownloadStreamOptions) (DownloadStreamResponse, error) {
@@ -409,7 +561,7 @@ func (b *Client) DownloadBuffer(ctx context.Context, buffer []byte, o *DownloadB
 	if o == nil {
 		o = &DownloadBufferOptions{}
 	}
-	return b.download(ctx, shared.NewBytesWriter(buffer), (downloadOptions)(*o))
+	return b.downloadBuffer(ctx, shared.NewBytesWriter(buffer), (downloadOptions)(*o))
 }
 
 // DownloadFile downloads an Azure blob to a local file.
@@ -448,7 +600,7 @@ func (b *Client) DownloadFile(ctx context.Context, file *os.File, o *DownloadFil
 	}
 
 	if size > 0 {
-		return b.download(ctx, file, *do)
+		return b.downloadFile(ctx, file, *do)
 	} else { // if the blob's size is 0, there is no need in downloading it
 		return 0, nil
 	}

sdk/storage/azblob/blob/constants.go

@@ -18,6 +18,8 @@ const (
 
 	// DefaultDownloadBlockSize is default block size
 	DefaultDownloadBlockSize = int64(4 * 1024 * 1024) // 4MB
+
+	defaultConcurrency = 5
 )
 
 // BlobType defines values for BlobType

sdk/storage/azblob/blockblob/chunkwriting.go

@@ -18,6 +18,7 @@ import (
 
 	"github.com/Azure/azure-sdk-for-go/sdk/azcore/streaming"
 	"github.com/Azure/azure-sdk-for-go/sdk/internal/uuid"
+	"github.com/Azure/azure-sdk-for-go/sdk/storage/azblob/internal/shared"
 )
 
 // blockWriter provides methods to upload blocks that represent a file to a server and commit them.
@@ -28,27 +29,8 @@ type blockWriter interface {
 	CommitBlockList(context.Context, []string, *CommitBlockListOptions) (CommitBlockListResponse, error)
 }
 
-// bufferManager provides an abstraction for the management of buffers.
-// this is mostly for testing purposes, but does allow for different implementations without changing the algorithm.
-type bufferManager[T ~[]byte] interface {
-	// Acquire returns the channel that contains the pool of buffers.
-	Acquire() <-chan T
-
-	// Release releases the buffer back to the pool for reuse/cleanup.
-	Release(T)
-
-	// Grow grows the number of buffers, up to the predefined max.
-	// It returns the total number of buffers or an error.
-	// No error is returned if the number of buffers has reached max.
-	// This is called only from the reading goroutine.
-	Grow() (int, error)
-
-	// Free cleans up all buffers.
-	Free()
-}
-
 // copyFromReader copies a source io.Reader to blob storage using concurrent uploads.
-func copyFromReader[T ~[]byte](ctx context.Context, src io.Reader, dst blockWriter, options UploadStreamOptions, getBufferManager func(maxBuffers int, bufferSize int64) bufferManager[T]) (CommitBlockListResponse, error) {
+func copyFromReader[T ~[]byte](ctx context.Context, src io.Reader, dst blockWriter, options UploadStreamOptions, getBufferManager func(maxBuffers int, bufferSize int64) shared.BufferManager[T]) (CommitBlockListResponse, error) {
 	options.setDefaults()
 
 	wg := sync.WaitGroup{}       // Used to know when all outgoing blocks have finished processing
@@ -265,49 +247,3 @@ func (ubi uuidBlockID) WithBlockNumber(blockNumber uint32) uuidBlockID {
 func (ubi uuidBlockID) ToBase64() string {
 	return blockID(ubi).ToBase64()
 }
-
-// mmbPool implements the bufferManager interface.
-// it uses anonymous memory mapped files for buffers.
-// don't use this type directly, use newMMBPool() instead.
-type mmbPool struct {
-	buffers chan mmb
-	count   int
-	max     int
-	size    int64
-}
-
-func newMMBPool(maxBuffers int, bufferSize int64) bufferManager[mmb] {
-	return &mmbPool{
-		buffers: make(chan mmb, maxBuffers),
-		max:     maxBuffers,
-		size:    bufferSize,
-	}
-}
-
-func (pool *mmbPool) Acquire() <-chan mmb {
-	return pool.buffers
-}
-
-func (pool *mmbPool) Grow() (int, error) {
-	if pool.count < pool.max {
-		buffer, err := newMMB(pool.size)
-		if err != nil {
-			return 0, err
-		}
-		pool.buffers <- buffer
-		pool.count++
-	}
-	return pool.count, nil
-}
-
-func (pool *mmbPool) Release(buffer mmb) {
-	pool.buffers <- buffer
-}
-
-func (pool *mmbPool) Free() {
-	for i := 0; i < pool.count; i++ {
-		buffer := <-pool.buffers
-		buffer.delete()
-	}
-	pool.count = 0
-}

sdk/storage/azblob/blockblob/chunkwriting_test.go

@@ -16,6 +16,7 @@ import (
 	"testing"
 	"time"
 
+	"github.com/Azure/azure-sdk-for-go/sdk/storage/azblob/internal/shared"
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
@@ -115,19 +116,19 @@ func calcMD5(data []byte) string {
 
 // used to track proper acquisition and closing of buffers
 type bufMgrTracker struct {
-	inner bufferManager[mmb]
+	inner shared.BufferManager[shared.Mmb]
 
 	Count int  // total count of allocated buffers
 	Freed bool // buffers were freed
 }
 
 func newBufMgrTracker(maxBuffers int, bufferSize int64) *bufMgrTracker {
 	return &bufMgrTracker{
-		inner: newMMBPool(maxBuffers, bufferSize),
+		inner: shared.NewMMBPool(maxBuffers, bufferSize),
 	}
 }
 
-func (pool *bufMgrTracker) Acquire() <-chan mmb {
+func (pool *bufMgrTracker) Acquire() <-chan shared.Mmb {
 	return pool.inner.Acquire()
 }
 
@@ -140,7 +141,7 @@ func (pool *bufMgrTracker) Grow() (int, error) {
 	return n, nil
 }
 
-func (pool *bufMgrTracker) Release(buffer mmb) {
+func (pool *bufMgrTracker) Release(buffer shared.Mmb) {
 	pool.inner.Release(buffer)
 }
 
@@ -161,7 +162,7 @@ func TestSlowDestCopyFrom(t *testing.T) {
 
 	errs := make(chan error, 1)
 	go func() {
-		_, err := copyFromReader(context.Background(), bytes.NewReader(bigSrc), fakeBB, UploadStreamOptions{}, func(maxBuffers int, bufferSize int64) bufferManager[mmb] {
+		_, err := copyFromReader(context.Background(), bytes.NewReader(bigSrc), fakeBB, UploadStreamOptions{}, func(maxBuffers int, bufferSize int64) shared.BufferManager[shared.Mmb] {
 			tracker = newBufMgrTracker(maxBuffers, bufferSize)
 			return tracker
 		})
@@ -270,7 +271,7 @@ func TestCopyFromReader(t *testing.T) {
 
 		var tracker *bufMgrTracker
 
-		_, err := copyFromReader(test.ctx, bytes.NewReader(from), fakeBB, test.o, func(maxBuffers int, bufferSize int64) bufferManager[mmb] {
+		_, err := copyFromReader(test.ctx, bytes.NewReader(from), fakeBB, test.o, func(maxBuffers int, bufferSize int64) shared.BufferManager[shared.Mmb] {
 			tracker = newBufMgrTracker(maxBuffers, bufferSize)
 			return tracker
 		})
@@ -322,7 +323,7 @@ func TestCopyFromReaderReadError(t *testing.T) {
 		reader: bytes.NewReader(make([]byte, 5*_1MiB)),
 		failOn: 2,
 	}
-	_, err := copyFromReader(context.Background(), &rf, fakeBB, UploadStreamOptions{}, func(maxBuffers int, bufferSize int64) bufferManager[mmb] {
+	_, err := copyFromReader(context.Background(), &rf, fakeBB, UploadStreamOptions{}, func(maxBuffers int, bufferSize int64) shared.BufferManager[shared.Mmb] {
 		tracker = newBufMgrTracker(maxBuffers, bufferSize)
 		return tracker
 	})

sdk/storage/azblob/blockblob/client.go

@@ -517,7 +517,7 @@ func (bb *Client) UploadStream(ctx context.Context, body io.Reader, o *UploadStr
 		o = &UploadStreamOptions{}
 	}
 
-	result, err := copyFromReader(ctx, body, bb, *o, newMMBPool)
+	result, err := copyFromReader(ctx, body, bb, *o, shared.NewMMBPool)
 	if err != nil {
 		return CommitBlockListResponse{}, err
 	}