透過您的圖書館登入
IP:216.73.216.202
  • 學位論文

基於雲端運算之即時影音串流轉碼服務平台之研製

Design and Implementation of Cloud-based Real-Time RTMP/RTSP Streaming Transcoder Service Platform

指導教授 : 黃能富

摘要


With the progress of network, multimedia applications are widely used in daily life. It is very convenient to use those multimedia applications. However there are still some problems we have to face. One of the problems is that distinct media services deliver media streaming based on different kinds of streaming protocols and codecs. And handheld devices may not support some protocols such as real time message protocol (RTMP) which is not supported by iPhone and iPad. In order to solve this problem, we design and implement a cloud-based real-time RTMP/RTSP streaming transcoder service platform. The system provides the service of changing real-time streaming protocols and codecs. And by utilizing virtual machines in cloud the scale of platform is not constrained by machine size. In addition, we also can control lots of virtual machines dynamically and efficiently. As the request of transcoding service grows up, it is almost impossible to control virtual machines manually. Hence, we design the auto-scaling algorithm to make machine manage machines come true. Besides, it also optimizes the cost of running the transcoder service platform. Thus with the proposed methods, in this thesis, we implement a transcoder service platform to play a role of bridge for heterogeneous media streaming.

關鍵字

cloud computing transcoder rtmp rtsp

並列摘要


隨著網際網路科技的發展,網路上許多多媒體影音應用軟體逐漸廣泛的應用在日常生活當中。使用這些軟體不只促進我們生活的便利同時也讓這些產品融入了我們的生活。但在這之中依然存在一些問題需要我們去解決。由不同家公司開發出來的多媒體伺服器使用的網路串流通訊協議以及影音編碼會不盡相同,而且有些網路串流通訊協議以及影音編碼手持式裝置並不支援。例如iPhone 和iPad 並不支援及時訊息協議(RTMP)。為了解決此類型問題建立起溝通的橋梁,我們設計實現了基於雲端運算之即時影音串流轉碼服務平台。此系統提供了及時影音串流的轉換通訊協議以及轉換影音編碼的服務。而且我們還利用雲端上的虛擬機器來建構我們的服務平台,如此一來系統規模將不會受到機器大小的限制隨時都可以加以擴充。除此之外,我們還可以隨意開關雲端上的大量虛擬機器,在需要的時候開啟,在不需要時就可以關閉減少浪費。隨著轉碼服務的需求增加,龐大的機器數量將不可能經由人工的方式進行操控,於是我們設計了一個自動調整系統規模的演算法。此方法實現了由機器管理機器的目標。除了自動管理之外,此系統還會針對租賃虛擬機器的花費做最佳化處理。透過上面所提到的方法,我們在這篇論文中提出並實現了一個扮演異質多媒體串流溝通橋梁的轉碼服務平台。

並列關鍵字

無資料

參考文獻


[3] Windows Media HTTP Streaming Protocol
Cloud Computing. UCB/EECS-2009-28, EECS Department, University of
state-of-the-art and research challenges. Journal of Internet Services and
Patterson, A. Rabkin, I. Stoica, and M. Zaharia, A view of cloud computing,!
[13] Buyya, R., Y. Chee Shin, et al. (2008). Market-Oriented Cloud Computing:

延伸閱讀