LASER: Light, Accurate Sharing dEtection and Repair

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 opa...

Full description

Saved in:
Bibliographic Details
Main Authors: Liang Luo, Sriraman, Akshitha, Fugate, Brooke, Shiliang Hu, Pokam, Gilles, Newburn, Chris J., Devietti, Joseph
Format: Conference Proceeding
Language:English
Subjects:
Architecture (computers)
Benchmark testing
Coherence
Construction
Hardware
Instruction sets
Lasers
Maintenance engineering
On-line systems
Operating systems
Radiation detectors
Repair
Run time (computers)
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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.
ISSN:2378-203X
DOI:10.1109/HPCA.2016.7446070