Student research poster: A low complexity cache sharing mechanism to address system fairness

Shared caches have become, de facto, the common design choice in current multi-cores, ranging from embedded devices to high-performance processors. In these systems, requests from multiple applications compete for the cache resources, degrading to different extents their progress, quantified as the...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2016 International Conference on Parallel Architecture and Compilation Techniques (PACT) S. 455
Hauptverfasser: Selfa, Vicent, Sahuquillo, Julio, Petit, Salvador, Gomez, Maria E.
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: ACM 01.09.2016
Schlagworte:
Binary trees
Complexity theory
Computers
Distance measurement
Performance evaluation
Program processors
Runtime
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Shared caches have become, de facto, the common design choice in current multi-cores, ranging from embedded devices to high-performance processors. In these systems, requests from multiple applications compete for the cache resources, degrading to different extents their progress, quantified as the performance of individual applications compared to isolated execution. The difference between the progresses of the running applications yields the system to unpredictable behavior and causes a fairness problem. This problem can be addressed by carefully partitioning cache resources among the contending applications, but to be effective, a partitioning approach needs to estimate the per-application progress.
DOI:10.1145/2967938.2971464