Design of new XOR-based hash functions for cache memories

Authors:
Sung-Jin Cho;Un-Sook Choi;Yoon-Hee Hwang;Han-Doo Kim
Affiliations:
Division of Mathematical Sciences, Pukyong National University, Busan 608-737, Republic of Korea;Department of Multimedia Engineering, Tongmyong University, Busan 608-711, Republic of Korea;Department of Information Security, Graduate School, Pukyong National University, Busan 608-737, Republic of Korea;School of Computer Aided Science, Inje University, Gimhae, 621-749, Republic of Korea
Venue:
Computers & Mathematics with Applications
Year:
2008

Citing 8
Cited 1

Pseudo-randomly interleaved memory

ISCA '91 Proceedings of the 18th annual international symposium on Computer architecture
A case for two-way skewed-associative caches

ISCA '93 Proceedings of the 20th annual international symposium on computer architecture
Cache Memories

ACM Computing Surveys (CSUR)
Skewed Associativity Improves Program Performance and Enhances Predictability

IEEE Transactions on Computers
XOR-Based Hash Functions

IEEE Transactions on Computers
Using Prime Numbers for Cache Indexing to Eliminate Conflict Misses

HPCA '04 Proceedings of the 10th International Symposium on High Performance Computer Architecture
Analysis of complemented CA derived from linear hybrid group CA

Computers & Mathematics with Applications
Behavior of complemented CA whose complement vector is acyclic in a linear TPMACA

Mathematical and Computer Modelling: An International Journal

Efficient address mapping of shared cache for on-chip many-core architecture

EuroPar'10 Proceedings of the 16th international Euro-Par conference on Parallel processing: Part I

Quantified Score

Hi-index	0.09

Visualization

Abstract

A hash function H is a computationally efficient function that maps bitstrings of arbitrary length to bitstrings of fixed length, called hash values. Hash functions have a variety of general computational uses. They are used in processors to augment the bandwidth of an interleaved multibank memory or to enhance the utilization of a prediction table or a cache. In this paper, we design new XOR-based hash functions, which compute each set index bit as XOR of a subset of the bits in the address by using the concepts of rank and null space. These are conflict-free hash functions which are of different types according to whether m is even or odd. To apply the constructed hash functions to the skewed-associative cache, we show that the degree of interbank dispersion between two hash functions is maximal.