Name
Abstract | ||
|---|---|---|
The trade-off between false sharing elimination and aggregation in distributed shared memory ( dsm ) systems has a major effect on their performance. Some studies in this area show that fine grain access is advantageous, while others advocate the use of large coherency units. One way to resolve the trade-off is to dynamically adapt the granularity to the application memory access pattern. In this paper, we propose a novel technique for implementing multiple sharing granularities over page-based dsm s. We present protocols for efficient switching between small and large sharing units during runtime. We show that applications may benefit from adapting the memory sharing to the memory access pattern, using both coarse grain sharing and fine grain sharing interchangeably in different stages of the computation. Our experiments show a substantial improvement in the performance using adapted granularity level over using a fixed granularity level. |
Year | DOI | Venue |
|---|---|---|
1999 | 10.1016/S0164-1212(00)00044-3 | Journal of Systems and Software |
Keywords | DocType | Volume |
fine grain access,multiple sharing granularity,coarse grain sharing,dsm systems,distributed shared memory,large sharing unit,dsm system,network programming,application memory access pattern,area show,fixed granularity level,dynamic adaptation,virtual parallel machine,false sharing elimination,granularity level,fine grain sharing interchangeably,protocols,false sharing,granularity,read only memory,memory management,message passing,computer integrated manufacturing,software systems,computer science,performance | Conference | 55 |
Issue | ISSN | ISBN |
1 | The Journal of Systems & Software | 0-7695-0350-0 |
Citations | PageRank | References |
2 | 0.39 | 15 |
Authors | ||
3 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ayal Itzkovitz | 1 | 172 | 13.22 |
| Nitzan Niv | 2 | 10 | 0.91 |
| Assaf Schuster | 3 | 2304 | 145.56 |