Fairness Metrics for Multithreaded Processors - Université de Rennes Accéder directement au contenu
Article Dans Une Revue IEEE Computer Architecture Letters Année : 2011

Fairness Metrics for Multithreaded Processors

Résumé

Multi-threaded processors execute multiple threads concurrently in order to increase overall throughput. It is well documented that multi-threading affects per-thread performance but, more importantly, some threads are affected more than others. This is especially troublesome for multi-programmed workloads. Fairness metrics measure whether all threads are affected equally. However defining equal treatment is not straightforward. Several fairness metrics for multi-threaded processors have been utilized in the literature, although there does not seem to be a consensus on what metric does the best job of measuring fairness. This paper reviews the prevalent fairness metrics and analyzes their main properties. Each metric strikes a different trade-off between fairness in the strict sense and throughput. We categorize the metrics with respect to this property. Based on experimental data for SMT processors, we suggest using the minimum fairness metric in order to balance fairness and throughput.
Fichier non déposé

Dates et versions

inria-00564560 , version 1 (09-02-2011)

Identifiants

Citer

Hans Vandierendonck, André Seznec. Fairness Metrics for Multithreaded Processors. IEEE Computer Architecture Letters, 2011, IEEE Computer Architecture Letters 2011, ⟨10.1109/L-CA.2011.1⟩. ⟨inria-00564560⟩
176 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More