Entropy representation of memory access characteristics and cache performance

Authors:
Takashi Yokota;Kanemitsu Ootsu;Takanobu Baba
Affiliations:
Utsunomiya University, Yoto, Utsunomiya, Tochigi, Japan;Utsunomiya University, Yoto, Utsunomiya, Tochigi, Japan;Utsunomiya University, Yoto, Utsunomiya, Tochigi, Japan
Venue:
ACST '08 Proceedings of the Fourth IASTED International Conference on Advances in Computer Science and Technology
Year:
2008

Citing 7
Cited 0

Compression-Based Program Characterization for Improving Cache Memory Performance

IEEE Transactions on Computers
Computer Architecture: A Quantitative Approach

Computer Architecture: A Quantitative Approach
Using Prime Numbers for Cache Indexing to Eliminate Conflict Misses

HPCA '04 Proceedings of the 10th International Symposium on High Performance Computer Architecture
Introducing entropies for representing program behavior and branch predictor performance

Proceedings of the 2007 workshop on Experimental computer science
On optimization of storage hierarchies

IBM Journal of Research and Development
Fractal nature of software-cache interaction

IBM Journal of Research and Development
Understanding prediction limits through unbiased branches

ACSAC'06 Proceedings of the 11th Asia-Pacific conference on Advances in Computer Systems Architecture

Quantified Score

Hi-index	0.00

Visualization

Abstract

Cache memories play important roles in modern computer architecture. Performance of a cache memory largely depends on spatial and temporal locality in memory access. Highly local access results good cache performance. In this paper, we discuss locality features in memory access and cache performance. We first discuss distribution function of memory access, and then, introduce an entropy concept. We define two new entropies based on memory access behavior and cache access. Preliminary evaluation results of cache performance and entropy values are reported.