TCE Guest Lecture: Optimized Networked Systems

Ymir Vigfusson (Reykjavik University)
Monday, 5.11.2012, 13:30
EE Meyer Building 861

Optimal use of computing resources is pivotal to minimize expenditures and improve competitiveness of companies. The computing technologies are continuously evolving and the users are adapting at a unparalleled pace, thus producing intriguing research questions.

In the first half of the talk, we observe that Internet service providers (ISPs) face surging Internet loads associated with real-time streaming video and various forms of dynamically generated, short-lived content. We propose a new system called GRADIENT aimed at reducing the load on providers of such and enabling scalable, bandwidth-sensitive streaming service for heterogeneous consumers. The core of the system is an overlay networking architecture intended to run directly on a content hosting platform, and which optimizes aggregate bandwidth use by transforming in-flight data to match the ideal stream quality – expressed as an economic utility – of the consuming client.

In the second half, we focus on improving resource consumption in the cloud. Web services use data caching to reduce end-user latency and database load. We highlight various challenges associated with supporting cloud-based caching services, in particular how to properly share these resources across cloud tenants. We also describe how these challenges were addressed by our prototype implementation, called Simple Cache for Cloud (SC2), and demonstrate the effectiveness of these techniques by experimentally evaluating our prototype on a synthetic multi-tenant workload.

The projects are done in collaboration with people at Technion TCE, Peking University, Cornell University, Reykjavik University and IBM Research Haifa.

Ymir Vigfusson is an Assistant Professor at the School of Computer Science at Reykjavik University. He received a B.Sc. in Mathematics from the University of Iceland (2005) and a Ph.D. in Computer Science from Cornell University (2009), where he researched ways to exploit group similarity and improve scalability in distributed systems. His dissertation was nominated for the ACM Doctoral Dissertation Award by Cornell. Before his appointment at Reykjavik University, Ymir was a post-doctoral scientist at IBM Research Haifa (2009-2011). Ymir's research projects include creating and optimizing systems and algorithms for distributed settings, and getting multicast and content distribution to work in a variety of environments. His work has been partially supported by a Fulbright Scholarship, a Yahoo! Research grant and a Grant-of-Excellence from the Icelandic Research Centre. In his spare time, Ymir plays the piano, dances ballroom and flies small .

Back to the index of events