Simulation as a tool for optimizing memory accesses on NUMA machines

Authors:
Jie Tao;Martin Schulz;Wolfgang Karl
Affiliations:
Institut für Rechnerentwurf und Fehlertoleranz, Universität Karlsruhe, Kaiserstr. 12, 76128 Karlsruhe, Germany;School of Electrical and Computer Engineering, Cornell University, Ithaca, NY 14853, USA;Institut für Rechnerentwurf und Fehlertoleranz, Universität Karlsruhe, Kaiserstr. 12, 76128 Karlsruhe, Germany
Venue:
Performance Evaluation - Performance modelling and evaluation of high-performance parallel and distributed systems
Year:
2005

Citing 12
Cited 0

A cache coherence approach for large multiprocessor systems

ICS '88 Proceedings of the 2nd international conference on Supercomputing
The SPLASH-2 programs: characterization and methodological considerations

ISCA '95 Proceedings of the 22nd annual international symposium on Computer architecture
Analyses and optimizations for shared address space programs

Journal of Parallel and Distributed Computing - Special issue on compilation techniques for distributed memory systems
Wisconsin Wind Tunnel II: A Fast, Portable Parallel Architecture Simulator

IEEE Concurrency
The Augmint multiprocessor simulation toolkit for Intel x86 architectures

ICCD '96 Proceedings of the 1996 International Conference on Computer Design, VLSI in Computers and Processors
An Automatic Iteration/Data Distribution Method Based on Access Descriptors for DSMM

LCPC '99 Proceedings of the 12th International Workshop on Languages and Compilers for Parallel Computing
Exploiting Spatial and Temporal Locality of Accesses: A New Hardware-Based Monitoring Approach for DSM Systems

Euro-Par '98 Proceedings of the 4th International Euro-Par Conference on Parallel Processing
A proposal for a new hardware cache monitoring architecture

Proceedings of the 2002 workshop on Memory system performance
Efficient memory simulation in SimICS

SS '95 Proceedings of the 28th Annual Simulation Symposium
OS Support for Improving Data Locality on CC-NUMA Compute Servers

OS Support for Improving Data Locality on CC-NUMA Compute Servers
Design and Analysis of Static Memory Management Policies for CC-NUMA Multiprocessors

Design and Analysis of Static Memory Management Policies for CC-NUMA Multiprocessors
Using complete machine simulation to understand computer system behavior

Using complete machine simulation to understand computer system behavior

Quantified Score

Hi-index	0.00

Visualization

Abstract

Due to the inherent non-uniformity in the memory system, programmers and users of non-uniform memory access (NUMA) machines have to take special care of the memory performance of their applications. This paper discusses a variety of potential improvements with respect to cache misses, cache invalidations, and inter-node communication. This study is based on the simulation tool SIMT, which models the memory hierarchy in detail and is capable of providing complete, accurate information about all dynamic memory references. This information can be used to analyze the memory access behavior of applications and thereby forms the basis for any optimization with respect to memory accesses.