- Abstract:
-
A DSM (Distributed-Shared Memory) cluster is an attractive parallel computing platform for scientific research as it provides programming advantages within a scalable and cost-effective hardware solution. However, the main drawback of DSM cluster is that it performs poorly on applications that require large shared memory because of the performance loss or overhead caused by memory locality. Consequently, an effective memory analysis methodology is mandatory to quantify the impact of such overhead. In this respect, a widely used technique is simulation. Nevertheless, both the rapid advance of DSM implementations and the intrinsically difficult design of cluster architectures impose a constraint on simulation model designs.
- Links To Paper
- No links available
- Bibtex format
- @InProceedings{EDI-INF-RR-0527,
- author = {
W. Marurngsith
and Roland Ibbett
},
- title = {dsimcluster: A Simulation Model for Efficient Memory Analysis Experiments of DSM Clusters},
- book title = {Procs of Summer Computer Simulation Conference},
- year = 2004,
- month = {Jul},
- }
|