Green Video-on-Demand (VOD) for Large-Scale Datacenters

Recent evidence suggests that global warming and climate change are primarily caused by greenhouse emissions from human activities. As a result, several “green” initiatives have been proposed to reduce carbon footprint. Our work in this area has focused primarily on green computing and communication in large scale datacenters where the cost to run and cool servers is becoming prohibitive. The main theme of our work has been smart disk storage/retrieval schemes that adapt resources to meet performance demands at lower overall energy cost.

In our work, we have focused on high-throughput applications such video-on-demand (or VOD), which are projected to account for over 90\% of internet traffic in the near future and will only exacerbate the energy issue in datacenters. Prior work in video streaming applications has shown that not all video data has the same visual importance. Put simply, some data can be lost or eliminated from the stream without heavily impacting the end user experience. In our research, we proposed an optimization scheme that reduces VOD server disk energy by retrieving only the data necessary to meet video quality constraints. For obtaining better performance, we also proposed novel data placement schemes for the server disks that resulted in better tradeoff between the quality delivered and energy overheads. The placement schemes followed two main principles: (i) allow faster and more energy efficient access for more popular videos; (ii) allow easier access to the data that can provide at least the minimum acceptable quality for all videos. Results showed that our approaches could decrease disk energy overheads and increase disk throughput with only a small degradation in video quality.

Our Conference and Journal Papers

NOTE: This directory contains pdf/ps files of articles that may be covered by copyright. You may browse the articles at your convenience, in the same spirit as you may read a journal or a proceedings article in a public library. Retrieving, copying, or distributing these files may violate copyright protection laws.

  • D. Forte, A. Srivastava, “Resource-Aware Architectures for Adaptive Particle Filter Based Visual Target Tracking”, ACM Transactions on Design Automation of Electronic Systems (TODAES), Vol. 18, No. 2, April 2013. [link]
  • D. Forte, A. Srivastava, “Energy-Aware and Quality-Scalable Data Placement and Retrieval for Disks in Video Server Environments”, IEEE International Conference on Computer Design (ICCD) , Oct. 2011. [link]
  • D. Forte, A. Srivastava, “Energy-Aware Video Storage and Retrieval in Server Environments”, International Green Computing Conference and Workshops (IGCC), July 2011. [link]