Proceedings Article10.1109/EURMIC.1998.711825
Cache probabilistic modeling for basic sparse algebra kernels involving matrices with a non-uniform distribution
Ramón Doallo,Basilio B. Fraguela,Emilio L. Zapata +2 more
- 25 Aug 1998
- Vol. 1, pp 10345
TL;DR: A probabilistic model to estimate the number of misses on a set associative cache with an LRU replacement algorithm is introduced and some new results focusing on different types of distributions that usually appear in some well-known real matrices suites, such as the Harwell-Boeing or NEP are presented.
read more
Abstract: A probabilistic model to estimate the number of misses on a set associative cache with an LRU replacement algorithm is introduced. Such modeling has been used by our group in previous work for sparse matrices with a uniform distribution of the non-zero elements. We present some new results focusing on different types of distributions that usually appear in some well-known real matrices suites, such as the Harwell-Boeing or NEP.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Efficient representation scheme for multidimensional array operations
TL;DR: A new scheme, extended Karnaugh map representation (EKMR), is proposed, to represent a multidimensional array by a set of two-dimensional arrays, and the experimental results show that the algorithms based on the EKMR scheme outperform thosebased on the TMR scheme and those provided by the Fortran 90 compiler.
60
Efficient data compression methods for multidimensional sparse array operations based on the EKMR scheme
TL;DR: The theoretical analysis and experimental results show that the ECRS/ECCS schemes are superior to the CRS/CCS schemes for all the evaluated criteria, except the space complexity in some case.
Efficient Data Parallel Algorithms for Multidimensional Array Operations Based on the EKMR Scheme for
Chun-Yuan Lin,Yeh-Ching Chung,Jen-Shiuh Liu +2 more
- 01 Jan 2003
TL;DR: It is shown that sequential multidimensional array operation algorithms based on the EKMR scheme have better performance than those based onThe traditional matrix representation (TMR) scheme.
Efficient data parallel algorithms for multidimensional array operations based on the EKMR scheme for distributed memory multicomputers
TL;DR: In this article, the authors proposed an extended Karnaugh map representation (EKMR) scheme for multidimensional array representation for distributed memory multicomputers and showed that the EKMR scheme has better performance than those based on the traditional matrix representation.
18
•Journal Article
Erratum: An Analytic Study of Caching in Computer Systems.
Duane Buck,Mukesh Singhal +1 more
TL;DR: In this paper, the independent reference model (IRM) of data reference in conjunction with the least recently used (LRU) cache block replacement policy is analyzed for the cache performance.
16
References
•Book
Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods
Richard Barrett
- 01 Jan 1987
TL;DR: In this book, which focuses on the use of iterative methods for solving large sparse systems of linear equations, templates are introduced to meet the needs of both the traditional user and the high-performance specialist.
Cache profiling and the SPEC benchmarks: a case study
Alvin R. Lebeck,Darien Wood +1 more
TL;DR: It is shown that cache profiling, using the CProf cache profiling system, improves program performance by focusing a programmer's attention on problematic code sections and providing insight into appropriate program transformations.
A Test Matrix Collection for Non-Hermitian Eigenvalue Problems
A. Bai,David Day,Jack Dongarra +2 more
- 01 Mar 1997
TL;DR: This document describes the mechanism for obtaining a copy of the matrices and for using the collection, and provides an easy access to "practical" eigenproblems for researchers, educators and students in the community.
116
Characterizing the behavior of sparse algorithms on caches
TL;DR: In this article, a methodology for modeling the irregular references of sparse codes using probabilistic methods is presented, and the behavior on cache of one of the most frequent primitives, SpMxV sparse matrix vector multiply, is analyzed.
An Analytic Study of Caching in Computer Systems
Duane Buck,Mukesh Singhal +1 more
TL;DR: Cache performance is analyzed for the independent reference model (IRM) of data reference in conjunction with the least recently used (LRU) cache block replacement policy to address cache coherence in parallel and distributed systems.
22