Understanding the Memory Behavior of Emerging Multi-core Workloads

Abstract

This paper characterizes the memory behavior on emerging RMS (recognition, mining, and synthesis) workloads for future multi-core processors. As multi-core processors proliferate across different application domains, and the number of on-die cores continues to increase, a key issue facing processor architects is the design of the on-die last level cache (LLC). In this paper, we explore the LLC design space for multi-threaded RMS workloads by examining the working set sizes, data sharing behavior, and spatial data locality. Our study reveals that these RMS workloads are memory intensive, have large working-set sizes greater than 16MB on average, exhibit a significant amount of data sharing, about47% on average, and show strong strided streaming access behavior with 77% of accesses in regular pattern. Based on the observations, we then investigate the potential cache architecture choices for future multi-core design. Our experiments show that for these workloads (a) large DRAM caches can be useful to address their large working sets; E.g., a 128MB DRAM cache can reduce the average L1 miss penalty by 18%; (b) shared last level cache provides better cache performance than private cache; E.g., a 8MB shared cache provides 25% performance improvement over a private one with the same total size; and (c) stride based hardware prefetcher provides significant performance benefit by 25%. As a result, we suggest a memory hierarchy with a 128MB DRAM cache, a 8MB on-die SRAM shared cache and an 8-entry stride prefetcher to accommodate RMS workloads.