Impact of Data Structure Layout on Performance

被引：10

作者：

Faria, Nuno ^{[1
]}

Silva, Rui ^{[1
]}

Sobral, Joao L. ^{[1
]}

机构：

[1] Univ Minho, CCTC DI, Braga, Portugal

来源：

PROCEEDINGS OF THE 2013 21ST EUROMICRO INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, AND NETWORK-BASED PROCESSING | 2013年

关键词：

locality; collections; multicore; !text type='Java']Java[!/text;

D O I：

10.1109/PDP.2013.24

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

One key issue to design parallel applications that scale on multicore systems is how to overcome the memory bottleneck. This paper presents a study of the impact of data structure layouts in locality of memory references, providing insights on strategies to ameliorate the memory bottleneck. The paper compares the performance of Java and C++ STL collections and presents the impact of locality of reference optimisations in a molecular dynamics simulation case study. The case study shows that the selected data structure layout has impact on single core performance, becoming a critical factor in the application scalability on multicore systems. Moreover, data collections provided in the Java language compromise performance due to pointer chasing and lack of spatial locality of memory references.

引用

页码：116 / 120

页数：5

共 7 条

[1] A portable programming interface for performance evaluation on modern processors [J].

Browne, S ;

Dongarra, J ;

Garner, N ;

Ho, G ;

Mucci, P .

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS, 2000, 14 (03) :189-204

[2]

Bull JM, 2000, CONCURRENCY-PRACT EX, V12, P375, DOI 10.1002/1096-9128(200005)12:6<375::AID-CPE480>3.0.CO

[3]

2-M

[4]

Hirzel M, 2007, PERF E R SI, V35, P265

[5]

Kumar S., 2006, IPDPS 06, P61

[6]

Wimmer C, 2008, INT SYM CODE GENER, P14

[7]

Yotov K, 2007, SPAA'07: PROCEEDINGS OF THE NINETEENTH ANNUAL SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES, P93

← 1 →