Talus: A simple way to remove cliffs in cache performance
Open Access
- 1 February 2015
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
Caches often suffer from performance cliffs: minor changes in program behavior or available cache space cause large changes in miss rate. Cliffs hurt performance and complicate cache management. We present Talus,1 a simple scheme that removes these cliffs. Talus works by dividing a single application's access stream into two partitions, unlike prior work that partitions among competing applications. By controlling the sizes of these partitions, Talus ensures that as an application is given more cache space, its miss rate decreases in a convex fashion. We prove that Talus removes performance cliffs, and evaluate it through extensive simulation. Talus adds negligible overheads, improves single-application performance, simplifies partitioning algorithms, and makes cache partitioning more effective and fair.Keywords
This publication has 45 references indexed in Scilit:
- Imbalanced cache partitioning for balanced data-parallel programsPublished by Association for Computing Machinery (ACM) ,2013
- CloudCache: Expanding and shrinking private cachesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- The ZCache: Decoupling Ways and AssociativityPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2010
- FlexDCPACM SIGOPS Operating Systems Review, 2009
- Producing wrong data without doing anything obviously wrong!Published by Association for Computing Machinery (ACM) ,2009
- Adaptive insertion policies for high performance cachingPublished by Association for Computing Machinery (ACM) ,2007
- Utility-Based Cache Partitioning: A Low-Overhead, High-Performance, Runtime Mechanism to Partition Shared Caches40th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO 2007), 2006
- Reconfigurable caches and their application to media processingPublished by Association for Computing Machinery (ACM) ,2000
- On-line construction of the convex hull of a simple polylineInformation Processing Letters, 1987
- Evaluation techniques for storage hierarchiesIBM Systems Journal, 1970