MLP-Aware Runahead Threads in a Simultaneous Multithreading Processor

Abstract. Threads experiencing long-latency loads on a simultaneous multith-reading (SMT) processor may clog shared processor resources without making forward progress, thereby starving other threads and reducing overall system throughput. An elegant solution to the long-latency load problem in SMT proces-sors is to employ runahead execution. Runahead threads do not block commit on a long-latency load but instead execute subsequent instructions in a speculative execution mode to expose memory-level parallelism (MLP) through prefetching. The key benefit of runahead SMT threads is twofold: (i) runahead threads do not clog resources on a long-latency load, and (ii) runahead threads exploit far-distance MLP. This paper proposes MLP-aware runahead threads: runahead execution is only initiated in case there is far-distance MLP to be exploited. By doing so, useless runahead executions are eliminated, thereby reducing the number of speculatively executed instructions (and thus energy consumption) while preserving the perfor-mance of the runahead thread and potentially improving the performance of the co-executing thread(s). Our experimental results show that MLP-aware runahead threads reduce the number of speculatively executed instructions by 13.9 % and 10.1 % for two-program and four-program workloads, respectively, compared to MLP-agnostic runahead threads while achieving comparable system throughput and job turnaround time.