An original-stream based solution for smoothly replaying high-definition videos in desktop virtualization systems

© 2014 by Knowledge Systems Institute Graduate School. How to smoothly replay high-definition (HD) videos in desktop virtualization systems has been a much needed yet challenging problem. Existing desktop virtualization systems apply well to classic office-applications but offer a very limited performance for multimedia applications, especially for replaying HD videos. In existing solutions, video is decoded on the server, and then decoded video data is highly compressed before being delivered to the client. However, high compression ratio requires a large amount of processing power and causes response delay, poor video quality and dropped frames that greatly deteriorate user experience. Although some solutions have been optimized for video replay, they are forced to modify media players and only support specific video formats. Therefore, we propose an original-stream based solution to provide good user experience for replaying HD videos in desktop virtualization systems without any modification on applications as well as support most of prevalent HD video formats. In our solution, video content is directly delivered to the client in its originally encoded state, and the actual video decoding and rendering are executed by the client’s GPU. The experimental results validate our method and show that this proposed approach measurably outperforms state-of-the-art solutions.