Efficient parallel processing on low-cost clusters with GAMMA active ports

被引:13
作者
Chiola, G [1 ]
Ciaccio, G [1 ]
机构
[1] Univ Genoa, DISI, I-16146 Genoa, Italy
关键词
fast messaging system; active messages; fast ethernet; multi-cast; collective communications; OS kernel; off-the-shelf components; linpack; parallel matrix multiplication; molecular dynamics simulation;
D O I
10.1016/S0167-8191(99)00108-8
中图分类号
TP301 [理论、方法];
学科分类号
081202 [计算机软件与理论];
摘要
The Genoa Active Message MAchine (GAMMA) is an efficient communication layer for 100base-T clusters of Personal Computers under the Linux operating system (OS), It is based on Active Ports, a communication mechanism derived from Active Messages. Active Ports share most of the low-level optimization opportunities with Generic Active Messages while offering a higher-level programming interface not only in the SPMD but also in the MIMD and client/server paradigms. In addition to point-to-point communications, multi-cast, barrier synchronization, scatter, and gather primitives have also been developed based on Active Ports and exploiting shared 100base-T LAN technology in an optimal way. GAMMA Active Ports deliver excellent communication performance at the user level (latency 13 mu s, maximum throughput 12.2 MByte/s, half-power point reached with 200 byte long messages), thus enabling cost-effective cluster computing on 100base-T. Despite being implemented at the kernel level in the Linux OS, performance numbers of GAMMA Active Ports are much better than many other LAN-oriented communication layers, including so called "user-level" ones (e.g. U-Net). Some code porting efforts have already shown that several applications are reasonably easy to develop on top of GAMMA and that they can actually take advantage of the efficient point-to-point as well as collective communication primitives offered by our prototype library implementation. A porting of the MPICH higher-level interface atop GAMMA is currently under way. (C) 2000 Published by Elsevier Science B.V. All rights reserved.
引用
收藏
页码:333 / 354
页数:22
相关论文
共 18 条
[1]
Chiola G, 1998, LECT NOTES COMPUT SC, V1362, P132, DOI 10.1007/BFb0052213
[2]
Chiola G, 1998, LECT NOTES COMPUT SC, V1470, P620, DOI 10.1007/BFb0057908
[3]
Implementing a low cost, low latency parallel platform [J].
Chiola, G ;
Ciaccio, G .
PARALLEL COMPUTING, 1997, 22 (13) :1703-1717
[4]
CHIOLA G, 1997, P 14 ITG GI C ARCH C
[5]
CHIOLA G, 1999, P 1999 INT C PAR DIS
[6]
Ciaccio G, 1998, LECT NOTES COMPUT SC, V1401, P884, DOI 10.1007/BFb0037221
[7]
Ciaccio G, 1998, LECT NOTES COMPUT SC, V1388, P524
[8]
Ciaccio G, 1998, LECT NOTES COMPUT SC, V1388, P534
[9]
CIACCIO G, 1999, THESIS U STUDI GENOV
[10]
CULLERD, 1994, GENERIC ACTIVE MESSA