TY - CONF
T1 - Shark
T2 - 2nd Symposium on Networked Systems Design and Implementation, NSDI 2005
AU - Annapureddy, Siddhartha
AU - Freedman, Michael J.
AU - Mazières, David
N1 - Funding Information:
Acknowledgments. We thank Vijay Karamcheti, Jinyuan Li, Robert Grimm, our shepherd, Peter Druschel, and members of NYU systems group for their helpful feedback on drafts of this p .p We ouldw like to thank Emulab (Robert Ricci, Timothy Stack, Leigh Sto,l and Jay Lepreau) and PlanetLab (Stve e Muir and Larry Peterson) researchers for assistance in running file-system xperimentse on their test-beds, as well as Eric Freudenthal and Jayanth Kumar Kannan for remote machine access. Fin ,l thanks to Jane-Ellen Long at USENIX for her consideration. This research asw conducted as part of the IRIS project (http://project-iris.net/), supported by the NSF under Cooperative Agreement No. ANI-0225660. Michael Freedman is supported by an NDSEG Fe - ship. David Mazi`s is supported by an Alfred P. Sloan Research Fe shi.
PY - 2005
Y1 - 2005
N2 - Network file systems offer a powerful, transparent interface for accessing remote data. Unfortunately, in current network file systems like NFS, clients fetch data from a central file server, inherently limiting the system's ability to scale to many clients. While recent distributed (peer-to-peer) systems have managed to eliminate this scalability bottleneck, they are often exceedingly complex and provide non-standard models for administration and accountability. We present Shark, a novel system that retains the best of both worlds-the scalability of distributed systems with the simplicity of central servers. Shark is a distributed file system designed for large-scale, wide-area deployment, while also providing a drop-in replacement for local-area file systems. Shark introduces a novel cooperative-caching mechanism, in which mutually-distrustful clients can exploit each others' file caches to reduce load on an origin file server. Using a distributed index, Shark clients find nearby copies of data, even when files originate from different servers. Performance results show that Shark can greatly reduce server load and improve client latency for read-heavy workloads both in the wide and local areas, while still remaining competitive for single clients in the local area. Thus, Shark enables modestly-provisioned file servers to scale to hundreds of read-mostly clients while retaining traditional usability, consistency, security, and accountability.
AB - Network file systems offer a powerful, transparent interface for accessing remote data. Unfortunately, in current network file systems like NFS, clients fetch data from a central file server, inherently limiting the system's ability to scale to many clients. While recent distributed (peer-to-peer) systems have managed to eliminate this scalability bottleneck, they are often exceedingly complex and provide non-standard models for administration and accountability. We present Shark, a novel system that retains the best of both worlds-the scalability of distributed systems with the simplicity of central servers. Shark is a distributed file system designed for large-scale, wide-area deployment, while also providing a drop-in replacement for local-area file systems. Shark introduces a novel cooperative-caching mechanism, in which mutually-distrustful clients can exploit each others' file caches to reduce load on an origin file server. Using a distributed index, Shark clients find nearby copies of data, even when files originate from different servers. Performance results show that Shark can greatly reduce server load and improve client latency for read-heavy workloads both in the wide and local areas, while still remaining competitive for single clients in the local area. Thus, Shark enables modestly-provisioned file servers to scale to hundreds of read-mostly clients while retaining traditional usability, consistency, security, and accountability.
UR - http://www.scopus.com/inward/record.url?scp=84936092287&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84936092287&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:84936092287
SP - 129
EP - 142
Y2 - 2 May 2005 through 4 May 2005
ER -