Where academic tradition
meets the exciting future

You are here: TUCS > PUBLICATIONS > Publication Search > Proactive Virtual Machine Allo...

Proactive Virtual Machine Allocation for Video Transcoding in the Cloud

Fareed Jokhio, Adnan Ashraf, Tewodros Deneke, Sebastien Lafond, Ivan Porres, Johan Lilius, Proactive Virtual Machine Allocation for Video Transcoding in the Cloud. In: Ivan Porres, Tommi Mikkonen, Adnan Ashraf (Eds.), Developing Cloud Software: Algorithms, Applications, and Tools, TUCS General Publication 60, 113–143, TUCS, 2013.

Abstract:

Video transcoding refers to the process of converting a digital video from one compressed format to another. It is a compute-intensive operation. Therefore, transcoding of a large number of simultaneous video streams requires a large-scale distributed system. Moreover, to handle different load conditions in a cost-efficient manner, the distributed system should be dynamically scalable. Infrastructure as a Service (IaaS) clouds currently offer computing resources, such as virtual machines (VMs), under the pay-per-use business model, which can be used to create a dynamically scalable cluster of video transcoding servers. Determining the number of VMs to provision for a dynamic cluster is still an open problem. In this chapter, we present a proactive VM allocation approach to scale video transcoding service on a given IaaS cloud. For better resource utilization, quality of service, and cost-efficiency, we use video segmentation at the Group of Pictures (GOP) level. The proposed approach is demonstrated in discrete-event simulations and an experimental evaluation involving different load patterns. We also present a prototype implementation of a distributed video transcoder based on the message passing programming model and a dynamic load balancing approach.

Files:

Full publication in PDF-format

BibTeX entry:

@INBOOK{cJoAsDeLaPoLi13a,
  title = {Proactive Virtual Machine Allocation for Video Transcoding in the Cloud},
  booktitle = {Developing Cloud Software: Algorithms, Applications, and Tools},
  author = {Jokhio, Fareed and Ashraf, Adnan and Deneke, Tewodros and Lafond, Sebastien and Porres, Ivan and Lilius, Johan},
  volume = {60},
  series = {TUCS General Publication},
  editor = {Porres, Ivan and Mikkonen, Tommi and Ashraf, Adnan},
  publisher = {TUCS},
  pages = {113–143},
  year = {2013},
  keywords = {Cloud computing, resource allocation, video transcoding},
}

Belongs to TUCS Research Unit(s): Embedded Systems Laboratory (ESLAB), Software Engineering Laboratory (SE Lab)

Edit publication