Abstract | ||
---|---|---|
Cloud computing systems consist of cloud client devices, which support various cloud computing services. As memory bottleneck degrades the performance of computing systems including cloud client devices, increasing the hit rates of last level caches (LLC) can improve the performance of cloud computing systems. To enhance the hit rates of LLC, we propose a new spatial locality aware prefetch technique. The proposed prefetch technique can fetch the data from main memory prior to actual requests to reduce the long latency to the main memory. To support the proposed technique, we introduce a new structure, LLC buffer which contains several memory blocks nearby the previously referenced memory block. In case that the LLC capacity is not enough, the proposed spatial locality-aware prefetch technique can improve the performance of cloud client devices significantly, especially for executing high spatial locality applications. Simulation results show that the proposed technique decreases LLC miss rates by up to 70.26 %, leading to performance improvement by up to 39.12 %. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1007/s10586-015-0470-8 | Cluster Computing |
Keywords | Field | DocType |
Last level cache,Spatial locality,LLC buffer,Prefetch,Cloud client device | Bottleneck,Locality,Latency (engineering),Computer science,Real-time computing,Memory management,Instruction prefetch,Computing systems,Operating system,Performance improvement,Cloud computing | Journal |
Volume | Issue | ISSN |
18 | 3 | 1386-7857 |
Citations | PageRank | References |
1 | 0.36 | 8 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hong Jun Choi | 1 | 30 | 5.74 |
Dong Oh Son | 2 | 21 | 4.19 |
Jong-Myon Kim | 3 | 91 | 25.99 |
Jinsul Kim | 4 | 81 | 25.54 |
Cheol Hong Kim | 5 | 73 | 24.39 |