Name
Abstract | ||
|---|---|---|
Efficient management of last level caches (LLCs) plays an important role in bridging the performance gap between processor cores and main memory. This paper is motivated by two key observations, based on our study of LLCs: 1) the capacity demand is highly non-uniform and dynamic at the set level, and 2) neither spatial nor temporal LLC management schemes, working separately as in prior work, can consistently and robustly deliver the best performance under different circumstances. Therefore, we propose a novel adaptive scheme, called STEM, which concurrently and dynamically manages both spatial and temporal dimensions of capacity demands at the set level. In the proposed scheme, a set-level monitor captures the temporal and spatial capacity demands of individual working sets and judiciously pairs off sets with complementary capacity demands so that the underutilized set in each pair can cooperatively cache the other's victim blocks. The controller also decides on the best temporal sharing patterns for the coupled sets in the event of inter-set space sharing. Further, if the LLC controller cannot find a complementary set for a particular set, STEM can still decide on the best set-level replacement policy for it. Our extensive execution-driven simulation data shows that the proposed scheme performs robustly and consistently well under various conditions. |
Year | DOI | Venue |
|---|---|---|
2010 | 10.1109/MICRO.2010.31 | IEEE Internet Computing |
Keywords | Field | DocType |
last level cache management,temporal sharing pattern,capacity demand,spatial capacity demand,intracore last level cache,llc controller,chip multiprocessors,microprocessor chips,particular set,cache storage,stem,last level caches,set-level non-uniformity of capacity demands,temporal llc management scheme,processor core,set level replacement policy,interset space sharing,spatial dimension,temporal dimension,spatiotemporal phenomena,cooperative caching,execution driven simulation data,cache management,proposed scheme,spatiotemporal management,complementary set,peer-to-peer computing,set level,complementary capacity demand,individual working set,measurement,robustness,resource management,electronics packaging,system on a chip,organizations | Resource management,Spatial capacity,Control theory,System on a chip,Computer science,Cache,Bridging (networking),Parallel computing,Robustness (computer science),Real-time computing,Multi-core processor | Conference |
ISSN | ISBN | Citations |
1072-4451 | 978-1-4244-9071-4 | 18 |
PageRank | References | Authors |
0.69 | 11 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Dongyuan Zhan | 1 | 40 | 2.91 |
| Hong Jiang | 2 | 2137 | 157.96 |
| Sharad C. Seth | 3 | 671 | 93.61 |