Departmental Papers (CIS)

Date of this Version

3-14-2016

Document Type

Conference Paper

Comments

2016 IEEE International Symposium on High Performance Computer Architecture (HPCA) was held March 12-16, 2016, in Barcelona, Spain.

Abstract

Contention for shared memory, in the forms of true sharing and false sharing, is a challenging performance bug to discover and to repair. Understanding cache contention requires global knowledge of the program's actual sharing behavior, and can even arise invisibly in the program due to the opaque decisions of the memory allocator. Previous schemes have focused only on false sharing, and impose significant performance penalties or require non-trivial alterations to the operating system or runtime system environment.

This paper presents the Light, Accurate Sharing dEtection and Repair (LASER) system, which leverages new performance counter capabilities available on Intel's Haswell architecture that identify the source of expensive cache coherence events. Using records of these events generated by the hardware, we build a system for online contention detection and repair that operates with low performance overhead and does not require any invasive program, compiler or operating system changes. Our experiments show that LASER imposes just 2% average runtime overhead on the Phoenix, Parsec and Splash2x benchmarks. LASER can automatically improve the performance of programs by up to 19% on commodity hardware.

Publication Source

2016 IEEE International Symposium on High Performance Computer Architecture (HPCA)

Start Page

261

Last Page

273

DOI

10.1109/HPCA.2016.7446070

Copyright/Permission Statement

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Keywords

false sharing, true sharing, cache contention, performance counters

Bib Tex

@INPROCEEDINGS{7446070,
author={L. Luo and A. Sriraman and B. Fugate and S. Hu and G. Pokam and C. J. Newburn and J. Devietti},
booktitle={2016 IEEE International Symposium on High Performance Computer Architecture (HPCA)},
title={LASER: Light, Accurate Sharing dEtection and Repair},
year={2016},
pages={261-273},
doi={10.1109/HPCA.2016.7446070},
month={March},}

Additional Files

laser-microbenchmarks.zip (446 kB)
Microbenchmarks used in Section 3.1

Share

COinS
 

Date Posted: 06 April 2016

This document has been peer reviewed.