H
Harsha V. Madhyastha
Researcher at University of Michigan
Publications - 101
Citations - 4795
Harsha V. Madhyastha is an academic researcher from University of Michigan. The author has contributed to research in topics: The Internet & Cloud computing. The author has an hindex of 31, co-authored 96 publications receiving 4374 citations. Previous affiliations of Harsha V. Madhyastha include University of California, Riverside & University of Washington.
Papers
More filters
Proceedings ArticleDOI
iPlane: an information plane for distributed services
Harsha V. Madhyastha,Tomas Isdal,Michael Piatek,Colin Dixon,Thomas Anderson,Arvind Krishnamurthy,Arun Venkataramani +6 more
TL;DR: The design, implementation, and evaluation of iPlane are presented, a scalable service providing accurate predictions of Internet path performance for emerging overlay services and demonstrating the feasibility and utility of the service by applying it to several representative overlay services in use today.
Proceedings Article
Improving the reliability of internet paths with one-hop source routing
TL;DR: This research demonstrates that one-hop source routing is easy to implement, adds negligible overhead, and achieves close to the maximum benefit available to indirect routing schemes, without the need for path monitoring, history, or a-priori knowledge of any kind.
Book ChapterDOI
FlowSense: monitoring network utilization with zero measurement cost
TL;DR: This paper proposes a push-based approach to performance monitoring in flow-based networks, where the network inform us of performance changes, rather than query it ourselves on demand, and discusses how the proposed passive approach can be combined with active approaches with low overhead.
Proceedings ArticleDOI
Moving beyond end-to-end path information to optimize CDN performance
Rupa Krishnan,Harsha V. Madhyastha,Sridhar Srinivasan,Sushant Jain,Arvind Krishnamurthy,Thomas Anderson,Jie Gao +6 more
TL;DR: This paper analyzes latencies measured from servers in Google's content distribution network (CDN) to clients all across the Internet to study the effectiveness of latency-based server selection, and finds that redirecting every client to the server with least latency does not suffice to optimize client latencies.
Proceedings ArticleDOI
SPANStore: cost-effective geo-replicated storage spanning multiple cloud services
TL;DR: SPANStore is presented, a key-value store that exports a unified view of storage services in geographically distributed data centers that can lower costs by over 10x in several scenarios, in comparison with alternative solutions that either use a single storage provider or replicate every object to every data center from which it is accessed.