從Azure下載BLOB

Question

從Azure存儲下載大型BLOB的最佳方式和最佳實踐是什麼？

• 我想用共享訪問簽名或共享訪問策略做的BLOB的本地副本。

• 估計BLOB大小爲500 MB到1 GB

• 互聯網連接從250KB改變爲4MB

Answer 1

在REST API的情況下使用HTTP request header

在存儲客戶端庫的情況下，對.NET使用DownloadRangeToStream

Answer 2

的最佳方式取決於你的情況。是否有一個特定的原因阻止您不按照以下（Java代碼）的方法做某件事？我已經看到過這樣的例子，但是有興趣瞭解爲什麼你認爲它不適合你。

public class SASread 
{ 
    public static void main(String[] args) throws URISyntaxException, FileNotFoundException, StorageException, IOException 
    {      
    URI baseuri = new URI("http://greengrass.blob.core.windows.net"); 
    CloudBlobClient blobclient = new CloudBlobClient(baseuri); 
    MyDownloadBlob("container1", 
      "sr=c&sv=2012-02-12&sig=ADsFalIpE1XkaneQpWwtpM3AhY%2BAwiUtbbo1ANbIoJA%3D&si=r", 
      blobclient);   
    } 

    public static void MyDownloadBlob(String containerName, String containerSAS, CloudBlobClient blobClient) throws URISyntaxException, StorageException, FileNotFoundException, IOException 
    { 
    String blobName = "image3.jpg"; 
    String localFileName = "c:\\myoutputimages\\image3.jpg"; 
    URI uri = new URI(blobClient.getEndpoint().toString() + "/" + 
         containerName + "/" + 
         blobName + 
         "?" + 
         containerSAS); 
    CloudBlockBlob sasBlob = new CloudBlockBlob(uri, blobClient); 
    File fileTarget = new File(localFileName); 
    sasBlob.download(new FileOutputStream(fileTarget)); 
    System.out.println("The blob at:\n" + uri + "\nwas downloaded from the cloud to local file:\n" + localFileName); 
    } 
}