Logo: to the web site of Uppsala University

uu.sePublikasjoner fra Uppsala universitet
Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Bundling: Reducing the Overhead of Multiprocessor Prefetchers
Uppsala universitet, Teknisk-naturvetenskapliga vetenskapsområdet, Matematisk-datavetenskapliga sektionen, Institutionen för informationsteknologi. Uppsala universitet, Teknisk-naturvetenskapliga vetenskapsområdet, Matematisk-datavetenskapliga sektionen, Institutionen för informationsteknologi, Datorteknik. Datorteknik.
Uppsala universitet, Teknisk-naturvetenskapliga vetenskapsområdet, Matematisk-datavetenskapliga sektionen, Institutionen för informationsteknologi. Uppsala universitet, Teknisk-naturvetenskapliga vetenskapsområdet, Matematisk-datavetenskapliga sektionen, Institutionen för informationsteknologi, Datorteknik. Datorteknik.
2004 (engelsk)Inngår i: 18th International Parallel and Distributed Processing Symposium: (IPDPS 2004), 2004Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

Prefetching has proven to be a useful technique for reducing cache misses in multiprocessors at the cost of increased coherence traffic. This is especially troublesome for snoop-based systems, where the available coherence bandwidth often is the scalability bottleneck. The bundling technique presented in this paper reduces the overhead caused by prefetching in two ways: piggybacking prefetches with normal requests, and requiring only one device to perform the snoop lookup for each prefetch transaction. This can reduce both the address bandwidth and the number of snoop lookups compared with a nonprefetching system. We describe bundling implementations for two important transaction types: reads and upgrades. While bundling could reduce the overhead of most existing prefetch schemes, the evaluation of bundling performed in this paper has been limited to two of them: sequential prefetching and Dahlgren´s adaptive sequential prefetching. Both schemes have their snoop bandwidth halved for all commercial and scientific benchmarks in the study. The combined effect of bundling applied to these prefetch schemes lowers the cache miss rate, the address bandwidth and the snoop bandwidth, compared with a system with no prefetching, for all applications. Bundling, will not reduce the data bandwidth introduced by a prefetch scheme. However, we argue that the data bandwidth is more easily scaled than the snoop bandwidth for snoop-based coherence systems.

Available as PDF (693 kB)

sted, utgiver, år, opplag, sider
2004.
Identifikatorer
URN: urn:nbn:se:uu:diva-72530OAI: oai:DiVA.org:uu-72530DiVA, id: diva2:100441
Tilgjengelig fra: 2005-05-25 Laget: 2005-05-25

Open Access i DiVA

Fulltekst mangler i DiVA

Person

Wallin, DanHagersten, Erik

Søk i DiVA

Av forfatter/redaktør
Wallin, DanHagersten, Erik
Av organisasjonen

Søk utenfor DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric

urn-nbn
Totalt: 515 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf