Abstract | ||
---|---|---|
As CMP platforms are widely adopted, more and more cores are integrated on to the die. To reduce the off-chip memory access, the last level cache is usually organized as a distributed shared cache. In order to avoid hot-spots, cache lines are interleaved across the distributed shared cache slices using a hash function. However, as we increase the number of cores and cache slices in the platform, this also implies that most of data references go to remote cache slices, thereby increasing the access latency significantly. In this paper, we propose a hybrid last level cache, which has some amount of private space and some amount of shared space on each cache slice. For workloads with no sharing, the goal is to provide more hits into the local slice while still keeping the overall miss rate low. For workloads with sufficient sharing, the goal is to allow more sharing in the last-level cache slice. We present hybrid last-level cache design options and study its hit/miss rate behavior for a number of important server applications and multi-programmed workloads. Our simulation results on running multi-programmed workloads based on SPEC CINT2000 as well as multithreaded workloads based on commercial server benchmarks (TPCC, SPECjbb, SAP and TPCE) show that this architecture is advantageous especially since it can improve the local hit rate significantly while keeping the overall miss rate similar to the shared cache. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1145/1399972.1399982 | SIGARCH Computer Architecture News |
Keywords | Field | DocType |
multi-programmed workloads,cache line,multithreaded workloads,last level cache,last-level cache slice,cache slice,shared cache,towards hybrid last level,remote cache slice,hybrid last-level cache design,private cache,hybrid cache,hybrid last level cache,chip,hot spot,hash function | Cache invalidation,Cache pollution,Cache,Computer science,Parallel computing,Cache algorithms,Page cache,Real-time computing,Cache coloring,Bus sniffing,Smart Cache,Operating system | Journal |
Volume | Issue | Citations |
36 | 2 | 17 |
PageRank | References | Authors |
0.98 | 16 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Li Zhao | 1 | 604 | 34.84 |
Ravishankar K. Iyer | 2 | 1119 | 75.72 |
Mike Upton | 3 | 110 | 4.32 |
Don Newell | 4 | 512 | 32.67 |