R. Manikantan, kaushik Rajan, and R. Govindarajan
Effective sharing of the last level cache has a significant influence on the overall performance of a multicore system. We observe that existing solutions control cache occupancy at a coarser granularity, do not scale well to large core counts and in some cases lack the flexibility to support a variety of performance goals. In this paper, we propose Probabilistic Shared CacheManagement (PriSM), a framework to manage the
cache occupancy of different cores at cache block granularity by controlling their eviction probabilities. The proposed framework requires only simple hardware changes to implement, can scale to larger core count and is flexible enough to support a variety of performance goals. We demonstrate the flexibility of PriSM, by plugging in the eviction probabilities needed to achieve goals like hit-maximization, fairness and QOS.
PriSM-HitMax improves performance by 18.4% over LRU and 11.5% over Vantage in a sixteen core machine. PriSM-Fairness improves fairness over existing solutions by 23.3% along with a performance improvement of 19.0%. PriSM-QOS successfully achieves the desired QOS targets.