Flexible Use of Memory for Replication/Migration in Cache-Coherent DSM Multiprocessors

DMCA

Flexible Use of Memory for Replication/Migration in Cache-Coherent DSM Multiprocessors (1998)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Vijayaraghavan Soundararajan , Mark Heinrich , Ben Verghese , Kourosh Gharachorloo , Anoop Gupta , John Hennessy

Venue:	In Proceedings of 25th International Symposium on Computer Architecture (ISCA
Citations:	37 - 4 self

BibTeX

@INPROCEEDINGS{Soundararajan98flexibleuse,
    author = {Vijayaraghavan Soundararajan and Mark Heinrich and Ben Verghese and Kourosh Gharachorloo and Anoop Gupta and John Hennessy},
    title = {Flexible Use of Memory for Replication/Migration in Cache-Coherent DSM Multiprocessors},
    booktitle = {In Proceedings of 25th International Symposium on Computer Architecture (ISCA},
    year = {1998},
    pages = {342--55}
}

OpenURL

Abstract

Given the limitations of bus-based multiprocessors, CC-NUMA is the scalable architecture of choice for shared-memory machines. The most important characteristic of the CC-NUMA architecture is that the latency to access data on a remote node is considerably larger than the latency to access local memory. On such machines, good data locality can reduce memory stall time and is therefore a critical factor in application performance. In this paper we study the various options available to system designers to transparently decrease the fraction of data misses serviced remotely. This work is done in the context of the Stanford FLASH multiprocessor. FLASH is unique in that each node has a single pool of DRAM that can be used in a variety of ways by the programmable memory controller. We use the programmability of FLASH to explore different options for cache-coherence and datalocality in compute-server workloads. First, we consider two protocols for providing base cache-coherence, one with centralized directory information (dynamic pointer allocation) and another with distributed directory information (SCI). While several commercial systems are based on SCI, we find that a centralized scheme has superior performance. Next, we consider different hardware and software techniques that use some or all of the local memory in a node to improve data locality. Finally, we propose a hybrid scheme that combines hardware and software techniques. These schemes work on the same base platform with both user and kernel references from the workloads. The paper thus offers a realistic and fair comparison of replication/migration techniques that has not previously been feasible. 1

Keyphrases

replication migration flexible use cache-coherent dsm multiprocessor software technique access data different option distributed directory information hybrid scheme memory stall time single pool various option good data locality system designer application performance base cache-coherence different hardware important characteristic stanford flash multiprocessor several commercial system access local memory fair comparison bus-based multiprocessor superior performance data locality scalable architecture cc-numa architecture compute-server workload base platform programmable memory controller local memory critical factor centralized scheme data miss remote node shared-memory machine replication migration technique dynamic pointer allocation centralized directory information

DMCA