Cloud Bursting Transcoder Nodes

Abstract: Within each web service there is a bottleneck which limits the maximum throughput. In this thesis the bottleneck of the media web service is in the task of transcoding. We explore the option of creating new transcoding nodes on-demand to deal with the computationally expensive task of video transcoding. The thesis presents a prototype for controlling the expansion of a transcoding service into the cloud. Using cloud infrastructure-as-a-service allows for creation of new nodes, without any modication to the software and via a prototype RESTful API using a simple syntax. The prototype also includes a systems integration framework (Chef) to configure new nodes to run any specified service. Due to licensing issues, we could not install the transcoding software itself while performing system testing. A substitute configuration of software of similar complexity was installed instead, and a large suite of tests shows that a single new node can be fully configured and a host notified about the node's existence within approximately 7 minutes. Updates/configurations can then be pushed out to all existing nodes, or a subset of nodes thereof. The destruction of nodes takes less than a minute and implicitly disassociate the nodes with all related resources, allowing for resource management. The thesis also explores the theoretical possibility of using MapReduce as a distributed model for a video transcoding service. Concepts in cloud computing, virtualization, video compression and transcoding are discussed as a part of an in-depth-study.