KIT | KIT-Bibliothek | Impressum | Datenschutz

Resource-conscious scheduling for energy efficiency on multicore processors

Merkel, Andreas; Stoess, Jan; Bellosa, Frank

Abstract:
In multicore systems, shared resources such as caches or the memory subsystem can lead to contention between applications running on different cores, entailing reduced performance and poor energy efficiency. The characteristics of individual applications, the assignment of applications to machines and execution contexts, and the selection of processor frequencies have a dramatic impact on resource contention, performance, and energy efficiency.
We employ the concept of task activity vectors for characterizing applications by resource utilization. Based on this characterization, we apply migration and co-scheduling policies that improve performance and energy efficiency by combining applications that use complementary resources, and use frequency scaling when scheduling cannot avoid contention owing to inauspicious workloads.
We integrate the policies into an operating system scheduler and into a virtualization system, allowing placement decisions to be made both within and across physical nodes, and reducing contention both for individual tasks and complete applications. Our evaluation based on the Linux operating system kernel an ... mehr

Open Access Logo


Volltext §
DOI: 10.5445/IR/1000026889
Originalveröffentlichung
DOI: 10.1145/1755913.1755930
Scopus
Zitationen: 103
Zugehörige Institution(en) am KIT Institut für Betriebs- und Dialogsysteme (IBDS)
Publikationstyp Proceedingsbeitrag
Jahr 2010
Sprache Englisch
Identifikator ISBN: 978-1-60558-577-2
urn:nbn:de:swb:90-268898
KITopen-ID: 1000026889
Erschienen in Proceedings of the 5th ACM SIGOPS EuroSys Conference (EurosSys'10), Paris, France, April 2010
Verlag ACM, New York (NY)
Seiten 14 S.
Nachgewiesen in Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page