Scalable cache-optimized concurrent FIFO queue for multicore architectures

Changwoo Min; Hyung Kook Jun; Won Tae Kim; Young Ik Eom

doi:10.1587/transinf.E95.D.2956

Scalable cache-optimized concurrent FIFO queue for multicore architectures

Changwoo Min, Hyung Kook Jun, Won Tae Kim, Young Ik Eom

Department of Computer Science and Engineering

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

A concurrent FIFO queue is a widely used fundamental data structure for parallelizing software. In this letter, we introduce a novel concurrent FIFO queue algorithm for multicore architecture. We achieve better scalability by reducing contention among concurrent threads, and improve performance by optimizing cache-line usage. Experimental results on a server with eight cores show that our algorithm outperforms state-ofthe- art algorithms by a factor of two.

Original language	English
Pages (from-to)	2956-2957
Number of pages	2
Journal	IEICE Transactions on Information and Systems
Volume	E95-D
Issue number	12
DOIs	https://doi.org/10.1587/transinf.E95.D.2956
State	Published - Dec 2012

Keywords

Cache-line contention
Compare-and-swap
Fetch-and-store
FIFO queue
Multicore processor

Access to Document

10.1587/transinf.E95.D.2956

Cite this

@article{cbe3f3a5d1ac4fb6965cbe540ed2bf13,

title = "Scalable cache-optimized concurrent FIFO queue for multicore architectures",

abstract = "A concurrent FIFO queue is a widely used fundamental data structure for parallelizing software. In this letter, we introduce a novel concurrent FIFO queue algorithm for multicore architecture. We achieve better scalability by reducing contention among concurrent threads, and improve performance by optimizing cache-line usage. Experimental results on a server with eight cores show that our algorithm outperforms state-ofthe- art algorithms by a factor of two.",

keywords = "Cache-line contention, Compare-and-swap, Fetch-and-store, FIFO queue, Multicore processor",

author = "Changwoo Min and Jun, \{Hyung Kook\} and Kim, \{Won Tae\} and Eom, \{Young Ik\}",

year = "2012",

month = dec,

doi = "10.1587/transinf.E95.D.2956",

language = "English",

volume = "E95-D",

pages = "2956--2957",

journal = "IEICE Transactions on Information and Systems",

issn = "0916-8532",

publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",

number = "12",

}

TY - JOUR

T1 - Scalable cache-optimized concurrent FIFO queue for multicore architectures

AU - Min, Changwoo

AU - Jun, Hyung Kook

AU - Kim, Won Tae

AU - Eom, Young Ik

PY - 2012/12

Y1 - 2012/12

N2 - A concurrent FIFO queue is a widely used fundamental data structure for parallelizing software. In this letter, we introduce a novel concurrent FIFO queue algorithm for multicore architecture. We achieve better scalability by reducing contention among concurrent threads, and improve performance by optimizing cache-line usage. Experimental results on a server with eight cores show that our algorithm outperforms state-ofthe- art algorithms by a factor of two.

AB - A concurrent FIFO queue is a widely used fundamental data structure for parallelizing software. In this letter, we introduce a novel concurrent FIFO queue algorithm for multicore architecture. We achieve better scalability by reducing contention among concurrent threads, and improve performance by optimizing cache-line usage. Experimental results on a server with eight cores show that our algorithm outperforms state-ofthe- art algorithms by a factor of two.

KW - Cache-line contention

KW - Compare-and-swap

KW - Fetch-and-store

KW - FIFO queue

KW - Multicore processor

UR - https://www.scopus.com/pages/publications/84870655159

U2 - 10.1587/transinf.E95.D.2956

DO - 10.1587/transinf.E95.D.2956

M3 - Article

AN - SCOPUS:84870655159

SN - 0916-8532

VL - E95-D

SP - 2956

EP - 2957

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

IS - 12

ER -

Scalable cache-optimized concurrent FIFO queue for multicore architectures

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this