首页出版物导航 用户协会 博客
 精确匹配 高级检索 检索历史
河北工业大学
详细信息

Towards Scalable Java HPC with Hybrid and Native Communication Devices in MPJ Express.

Title: Towards Scalable Java HPC with Hybrid and Native Communication Devices in MPJ Express.
Authors: Javed, Ansar1 ansar.javed@seecs.edu.pk
Qamar, Bibrak2 bibrak.qamar@seecs.nust.edu.pk
Jameel, Mohsan mohsan.jameel@ismll.uni-hildesheim.de
Shafi, Aamir2 aamir.shafi@seecs.nust.edu.pk
Carpenter, Bryan3 bryan.carpenter@port.ac.uk
Source: International Journal of Parallel Programming. Dec2016, Vol. 44 Issue 6, p1142-1172. 31p.
Document Type: Article
Subject Terms: OBJECT-oriented programming languages
HIGH performance computing
ELECTRONIC data processing
JAVA (Computer program language)
COMPUTATIONAL grids (Computer systems)
Author-Supplied Keywords: High Performance Computing in Java ; Hybrid parallelism in Java ; MPJ Express ; Native communication device for Java ; Parallel computing
NAICS/Industry Codes: 518210 Data Processing, Hosting, and Related Services
Abstract: MPJ Express is a messaging system that allows application developers to parallelize their compute-intensive sequential Java codes on High Performance Computing clusters and multicore processors. In this paper, we extend MPJ Express software to provide two new communication devices. The first device-called hybrid-enables MPJ Express to exploit hybrid parallelism on cluster of multicore processors by sitting on top of existing shared memory and network communication devices. The second device-called native-uses JNI wrappers in interfacing MPJ Express to native MPI implementations like MPICH and Open MPI. We evaluate performance of these devices on a range of interconnects including 1G/10G Ethernet, 10G Myrinet and 40G InfiniBand. In addition, we analyze and evaluate the cost of MPJ Express buffering layer and compare it with the performance numbers of other Java MPI libraries. Our performance evaluation reveals that the native device allows MPJ Express to achieve comparable performance to native MPI libraries-for latency and bandwidth of point-to-point and collective communications-which is a significant gain in performance compared to existing communication devices. The hybrid communication device-without any modifications at application level-also helps parallel applications achieve better speedups and scalability by exploiting multicore architecture. Our performance evaluation quantifies the cost incurred by buffering and its impact on overall performance of software. We witnessed comparative performance as both new devices improve application performance and achieve upto 90 % of the theoretical bandwidth available without application rewriting effort-including NAS Parallel Benchmarks, point-to-point and collective communication. [ABSTRACT FROM AUTHOR]
(Copyright applies to all Abstracts.)
Author Affiliations: 1Mohammad Ali Jinnah University (MAJU) , Islamabad Pakistan ; 2SEECS, National University of Sciences and Technology (NUST) , Islamabad Pakistan ; 3School of Computing , University of Portsmouth , Portsmouth UK
ISSN: 0885-7458 (Sherpa/RoMEO, JCR)
PageCount: 1142-1173
volume: 44
issue: 6
issn: 08857458
pubdate: 2016
DOI: https://doi.org/10.1007/s10766-015-0375-4
JCR分区:
ESI学科:
最新一年影响因子:
当年影响因子: