Using idle workstations to implement predictive prefetching

Simple item page
dc.contributor.authorWang, JYQ
dc.contributor.authorOng, JS
dc.contributor.authorCoady, Y
dc.contributor.authorFeeley, MJ
dc.date.accessioned2009-07-31T15:47:28Z
dc.date.available2009-07-31T15:47:28Z
dc.date.copyright2000en
dc.date.issued2000
dc.description©2000 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.en
dc.description.abstractThe benefits of Markov-based predictive prefetching have been largely overshadowed by the overhead required to produce high-quality predictions. While both theoretical and simulation results for prediction algorithms appear promising, substantial limitations exist in practice. This outcome can be partially attributed to the fact that practical implementations ultimately make compromises in order to reduce overhead. These compromises limit the level of algorithm complexity, the variety of access patterns and the granularity of trace data that the implementation supports. This paper describes the design and implementation of GMS-3P (Global Memory System with Parallel Predictive Prefetching), an operating system kernel extension that offloads prediction overhead to idle network nodes. GMS-3P builds on the GMS global memory system, which pages to and from remote workstation memory. In GMS-3P, the target node sends an online trace of an application's page faults to an idle node that is running a Markov-based prediction algorithm. The prediction node then uses GMS to prefetch pages to the target node from the memory of other workstations in the network. Our preliminary results show that predictive prefetching can reduce the remote-memory page fault time by 60% or more and that, by offloading prediction overhead to an idle node, GMS-3P can reduce this improved latency by between 24% and 44%, depending on the Markov model order.en
dc.identifier.citationWang, J.Y.Q.; Ong, J.S.; Coady, Y.; Feeley, M.J., High-Performance Distributed Computing, 2000. Proceedings. The Ninth International Symposium onen
dc.identifier.urihttp://hdl.handle.net/1828/1487
dc.language.isoenen
dc.publisherIEEEen
dc.subjectaccess patternsen
dc.subjectalgorithm complexityen
dc.subjectapplication page faultsen
dc.subjectcompromisesen
dc.subjectdistributed memory systemsen
dc.subjectglobal memory systemen
dc.subjectGMS-3Pen
dc.subjecthigh-quality predictionsen
dc.titleUsing idle workstations to implement predictive prefetchingen
dc.typeArticleen

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Using idle workstations to implement predictive prefetching.pdf
Size:
712.9 KB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.83 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Faculty Publications (Engineering and Computer Science)
 
University of Victoria Libraries

We acknowledge and respect the Lək̓ʷəŋən (Songhees and Esquimalt) Peoples on whose territory the university stands, and the Lək̓ʷəŋən and W̱SÁNEĆ Peoples whose historical relationships with the land continue to this day.