DMCA
MLP-Aware Runahead Threads in a Simultaneous Multithreading Processor
Cached
Download Links
BibTeX
@MISC{Craeynest_mlp-awarerunahead,
author = {Kenzo Van Craeynest and Stijn Eyerman and Lieven Eeckhout},
title = {MLP-Aware Runahead Threads in a Simultaneous Multithreading Processor},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract. Threads experiencing long-latency loads on a simultaneous multithreading (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 processors is to employ runahead execution. Runahead threads do not block commit on a longlatency 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 performance 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. 1
Keyphrases
mlp-aware runahead thread runahead thread simultaneous multithreading processor far-distance mlp long-latency load runahead execution energy consumption runahead smt thread speculative execution mode useless runahead execution four-program workload co-executing thread memory-level parallelism processor resource overall system throughput mlp-agnostic runahead thread execute subsequent instruction longlatency load smt processor key benefit simultaneous multithreading experimental result comparable system throughput long-latency load problem job turnaround time elegant solution
