C# version: http://code.msdn.microsoft.com/CSMultiThreadedWebDownloade-2be7db9b
VB version: http://code.msdn.microsoft.com/VBMultiThreadedWebDownloade-57e5475b
Today’s code sample demonstrate how to to download a file using multiple threads. It’s a multi-threaded web downloaded created in C# and VB.NET!
The sample was written by our star sample developer: Ruiz Yi.
You can find more code samples that demonstrate the most typical programming scenarios by using Microsoft All-In-One Code Framework Sample Browser or Sample Browser Visual Studio extension. They give you the flexibility to search samples, download samples on demand, manage the downloaded samples in a centralized place, and automatically be notified about sample updates. If it is the first time that you hear about Microsoft All-In-One Code Framework, please watch the introduction video on Microsoft Showcase, or read the introduction on our homepage http://1code.codeplex.com/.
The sample demonstrates how to download a file on Web using multiple threads.
The class System.Net.WebClient has a DownloadProgressChanged event, and you can register this event to show the progress. But this class does not support Pause/Resume.
The class MultiThreadedWebDownloader in this sample could be used to download data through internet and supports following features:
1. Download the whole file using multiple threads.
2. Set the proxy.
3. Set the buffer and cache size.
4. Start, Pause, Resume and Cancel a download.
5. Supply the file size, download speed and used time.
6. Expose the events StatusChanged, DownloadProgressChanged and DownloadCompleted.
1. To pause and resume a download, or download a file using multi-threads, the server must support "Accept-Ranges" header, so that we can add range to the webrequset to download a specified block of the file.
2. If you are trying to download a single file using multiple threads/ multiple HttpWebRequest’s, multiple requests (each asking for a different range) will all be pipelined over a single connection (not necessarily all). You have to take into consideration as to how many threads you are spinning, and how many max Connections you have set for the ServicePointManager. By default, ServicePointManager has a default of 2 max connections for console applications, so only 2 requests can be active at a time and will be downloading 2 parts of the large file. So yes, your approach will speed up the download, but you also need to take into account the maximum number of connections. If you increase the number of available connections, you will get a better download rate. But, do not increase it to a very large value, otherwise you will end up spiking the CPU due to the overhead of creating multiple connections simultaneously. Try to limit the max Conn number = 12 * # CPU which asp.net uses.
Step1. Build the sample project in Visual Studio 2010.
Step2. Run VBMultiThreadedWebDownloader.exe.
Step3. Type following link as url.
Step4. Click the button "Check Url". Then the textbox "Url" and the button "Check Url" will be disabled, the textbox "Local path" and the button "Download" will be enabled.
Step5. Type a local path like D:\DotNetFx4.exe.
Step6. Click the button "Download", you will see the status "Downloading" and the summary "Received: ***KB, Total: ***KB, Speed: ***KB/s Thread: *", the progressbar will also change. In Windows Explorer, you will find a file D:\DotNetFx4.exe.tmp.
Step7. Click the button "Pause", you will see the status "Paused" and the summary "Received: ***KB, Total: ***KB, Time: ***", the progressbar will also stop.
Step8. Click the button "Resume", you will see the status "Downloading" and the summary "Received: ***KB, Total: ***KB, Speed: ***KB/s", the progressbar will also change.
Step9. When the download completes, you will see the status "Completed" and the summary "Received: ***KB, Total: ***KB, Time: ***", the progressbar will also stop. In Windows Explorer, you will find a file D:\DotNetFx4.exe.
Using the Code
1. When the application starts, it loads the proxy settings in app.config.
2. Before downloading the file, MultiThreadedWebDownloader has to check whether
2.1 The file exists.
2.2 The remote server supports "Accept-Ranges" header.
2.3 The size of the file is more than 0.
3. If the check is passed, then user can specify the local path and start the download.
When the download starts, check whether the destination file exists. If not, create a local file with the same size as the file to be downloaded.
If the remote server does not support "Accept-Ranges" header, the downloader will create just one HttpDownloadClient to download the file in a single thread. Else, the downloader will create multiple HttpDownloadClients to download the file in multiple threads.
4. When a HttpDownloadClient object is created, initialize the fields/properties StartPoint, EndPoint, BufferSize, MaxCacheSize, BufferCountPerNotification, Url
DownloadPath and Status.
5. Each download thread will read a buffer of bytes from the response stream, and store the buffer to a MemoryStream cache first. If the cache is full, or the download is paused, canceled or completed, write the data in cache to local file.
6. Raise the event DownloadProgressChanged when read a specified number of buffers.
7. If the download is paused, each HttpDownloadClient will store the downloaded size. When it is resumed, start to download the file from a start point.
8. Update the used time and status when the current download stops.
9. Raise the event DownloadCompleted when the download is completed or canceled.
How can i check if file download completed