ACCELERATING CRITICAL SECTION EXECUTION WITH ASYMMETRIC MULTICORE ARCHITECTURES

被引：11

作者：

Suleman, M. Aater ^{[1
]}

Mutlu, Onur ^{[2
]}

Qureshi, Moinuddin K. ^{[3
]}

Patt, Yale N. ^{[1
]}

机构：

[1] Univ Texas Austin, Austin, TX 78712 USA

[2] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA

[3] IBM Res Corp, Yorktown Hts, NY USA

来源：

IEEE MICRO | 2010年 / 30卷 / 01期

关键词：

CMP; Critical sections; Heterogeneous cores; Locks; Multicore; Parallel programming; Serialization;

D O I：

10.1109/MM.2010.7

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Contention for critical sections can reduce performance and scalability by causing thread serialization. The proposed accelerated critical sections mechanism reduces this limitation acs executes critical sections on the high-performance core of an asymmetric chip multiprocessor (acmp), which can execute them faster than the smaller cores can.

引用

页码：60 / 70

页数：11

共 18 条

[1]

AMDAHL GM, 1967, FEDERATION INFORM PR, P483

[2]

ANNAVARAM M, 2005, SIGARCH COMPUT ARCHI, V33, P298

[3] IMPLEMENTING REMOTE PROCEDURE CALLS [J].

BIRRELL, AD ;

NELSON, BJ .

ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1984, 2 (01) :39-59

[4]

Culler D., 1998, Parallel Computer Architecture: A Hardware/software Approach

[5]

HERLIHY M., 1993, P 20 ANN INT S COMPU, P289, DOI DOI 10.1145/165123.165164.URL

[6] Amdahl's law in the multicore era [J].

Hill, Mark D. ;

Marty, Michael R. .

COMPUTER, 2008, 41 (07) :33-+

[7]

Ipek Engin., 2007, PROC 34 ANN INT S CO, P186

[8] Heterogeneous chip multiprocessors [J].

Kumar, R ;

Tullsen, DM ;

Jouppi, NP ;

Ranganathan, P .

COMPUTER, 2005, 38 (11) :32-+

[9] BRANCH-AND-BOUND METHODS - A SURVEY [J].

LAWLER, EL ;

WOOD, DE .

OPERATIONS RESEARCH, 1966, 14 (04) :699-+

[10]

Martinez-Miranda J., 2002, P 3 WORKSH AG BAS SI, P18

← 1 2 →