建置於私有雲之具有備援機制的視訊轉碼系統

由於現今連網設備軟硬體效能的增進與網際網路的快速發展，使用者對於多媒體串流服務的需求日漸增長，衍生大量需要密集運算的多媒體串流訊息，如果使用傳統單一伺服器將會有運算能力的極限。如能採用無所不在、取用資源近乎無限制的雲端運算，可加快多媒體串流服務的運算速度。針對多媒體串流服務中視訊轉碼的服務需求，本論文基於私有雲的架構，設計並實作相關的佈署模型，以建構 IaaS 服務模型的開源計畫 OpenNebula 為核心，實現具多台基礎設備的雲端視訊轉碼系統，轉碼器的解碼端選擇使用將編碼端的複雜度轉移到解碼端上的分散式視訊編碼，在雲端系統中處理其複雜度較高的解碼端，如此影像上傳者即可使用較為低階處理能力的設備。而為提供影像給各種異質設備與環境的使用者，轉碼器將視訊資料轉碼為具多種適應性的可調性視訊編碼給使用者觀看。而為提升此系統的可靠性，在設計時加入了備援機制，可在機器發生故障時將運算需求轉移至其它正常運作的機器上。所設計的雲端視訊轉碼系統在以建置系統滿載的狀況下，與單機運行的轉碼器相比，平均可以加快7.24倍的轉碼時間，轉碼速率可達7.28 FPS。

關鍵字

轉碼器；可調性視訊編碼；分散式視訊編碼；雲端運算；備援機制

並列摘要

Because of the rapid development of the internet and the enhancement of software and hardware networking equipment today, the multimedia stream service demands of user have grown increasingly. Therefore, intensive multimedia streaming information computing requirement would be limited if the traditional architecture withsingle-server system is applied. Fortunately, the powerful cloud computing can solve these high-complexity tasks. We will focus on the cloud-based transcoding services to speed up the transcoding speed of multimedia stream service. In this thesis, the design and implement relevant deployment model are based on the open source toolkit “OpenNebula”. This tool can help us to implement cloud video transcoding system with multiple servers and realize implementations of infrastructure as a service (IaaS) . We consider Distributed Video Coding (DVC) stream as the input of the transcoder, because DVC can transfer the high-complexity computation from the encoder to the decoder. With cloud computing, the complicated DVC decoder can be fulfilled easily. So the client can record and upload the video content with low-level processing equipment. According to various download scenarios, Scalable Video Coding is considered as the encoder in the transcoder to meet the varying requests in quality, resolution and frame-rate. To enhance the system reliability, we introduce backup and recovery mechanism into the cloud video transcoding system. Therefore, which can transfer the tasks from the servers with faults to the normal ones. The proposed cloud video transcoding system can increase the transcoding speed up to 7.28 frames per second. The improvement is about 7.24 times in average compared with traditional single-server transcoding.

並列關鍵字

Transcoder ； SVC ； DVC ； Cloud Computing ； Machine-failure Recovery

參考文獻

[1]H. Schwarz, D. Marpe, T. Wiegand,“Overview of the Scalable Video Coding Extension of the H.264/AVC Standard,” IEEE Trans. on Circuits and Systems for Video Technology, Vol.17, No. 9, pp.1013-1120, Sept. 2007.

[2]C.A. Segall, G.J. Sullivan, “Spatial Scalability Within the H.264/AVC Scalable Video Coding Extension,” IEEE Trans. on Circuits and Systems for Video Technology, Vol.17, No. 9, pp. 1121–1135, Sep. 2007.

[3]H. Schwarz, M. Wien, “The Scalable Video Coding Extension of The H.264/AVC Standard,” IEEE Signal Processing Magazine, Vol. 25, No. 2, pp. 135~141, March 2008.

[4]D. Slepian and J. K. Wolf, “Noiseless coding of correlated information source,” IEEE Trans. Inform. Theory, Vol. 19, No. 4, pp. 471-480, July 1973.

[5]Wyner and J. Ziv, “The rate-distortion function for source with side information at the decoder,” IEEE Trans. Inform. Theory, Vol. 22, No.1, pp. 1-10, Jan 1976.

國際替代計量

建置於私有雲之具有備援機制的視訊轉碼系統

未授權

主題瀏覽