Cache Replacement for Multi-Threaded Applications Using Context Based Data Pattern Exploitation Technique

Abstract

The impact of various cache replacement policies act as the main deciding factor of system performance and efficiency in Chip Multi-Core Processors (CMP). Many existing cache replacement polices such as the Least Recently Used (LRU), Most Recently Used (MRU), Not Recently Used (NRU) etc. have proved to work well in the shared L2 cache for most of the data set patterns generated by current applications. But when it comes to parallel multi-threaded applications which generate differing patterns of workload at different intervals, the above specified schemes might prove sub-optimal as they generally do not abide by the spatial and temporal locality theories. This paper proposes a novel cache replacement policy that is targeted towards such applications. Context Based Data Pattern Exploitation Technique (CB-DPET) assigns a counter for every block of the L2 cache. It then closely monitors the data access patterns of various threads and modifies the counter values appropriately to maximize the overall hit percentage. Experimental results obtained by using the PARSEC benchmarks have shown an average improvement of 8% to 9% in overall hits at L2 cache level when compared to the conventional LRU algorithm.