10.1184/R1/6619859.v1
Michael Abd-El-Malek
Michael
Abd-El-Malek
William V. Courtright II
William V.
Courtright
Chuck Cranor
Chuck
Cranor
Gregory R. Ganger
Gregory R.
Ganger
James Hendricks
James
Hendricks
Andrew J. Klosterman
Andrew J.
Klosterman
Michael Mesnier
Michael
Mesnier
Manish Prasad
Manish
Prasad
Brandon Salmon
Brandon
Salmon
Raja R. Sambasivan
Raja R.
Sambasivan
Shafeeq Sinnamohideen
Shafeeq
Sinnamohideen
John D. Strunk
John D.
Strunk
Eno Thereska
Eno
Thereska
Matthew Wachs
Matthew
Wachs
Jay J. Wylie
Jay J.
Wylie
Ursa Minor: Versatile Cluster-based Storage (CMU-PDL-05-104)
Carnegie Mellon University
2005
Data Storage
2005-12-01 00:00:00
Journal contribution
https://kilthub.cmu.edu/articles/journal_contribution/Ursa_Minor_Versatile_Cluster-based_Storage_CMU-PDL-05-104_/6619859
No single encoding scheme or fault model is optimal for all data. A versatile storage system allows them to be matched to access patterns, reliability requirements, and cost goals on a per-data item basis. Ursa Minor is a cluster-based storage system that allows data-specific selection of, and on-line changes to, encoding schemes and fault models. Thus, different data types can share a scalable storage infrastructure and still enjoy specialized choices, rather than suffering from “one size fits all.” Experiments with Ursa Minor show performance benefits of 2–3× when using specialized choices as opposed to a single, more general, configuration. Experiments also show that a single cluster supporting multiple workloads simultaneously is much more efficient when the choices are specialized for each distribution rather than forced to use a “one size fits all” configuration. When using the specialized distributions, aggregate cluster throughput nearly doubled.