Exploiting Gigabit Ethernet capacity for cluster applications

作者: G. Ciaccio , M. Ehlert , B. Schnor

DOI: 10.1109/LCN.2002.1181844

关键词:

摘要: In this paper we report about the recently completed porting of GAMMA to Netgear GA621 Gigabit Ethernet adapter, and provide a comparison among GAMMA, MPI/GAMMA, TCP/IP MPICH/TCP, based on older GA620 network adapters using different device drivers, in cluster PC running Linux 2.4. (the Genoa Active Message Machine) is lightweight messaging system an active message-like paradigm, originally designed for efficient exploitation Fast interconnects. The includes simple latency/bandwidth evaluation systems both adapters, as well performance comparisons NAS Parallel Benchmarks end-user fluid dynamics application called Modular Ocean Model (MOM). analysis results provides useful hints concerning use with clusters PC. particular, it emerges that combination low end-to-end latency (8.5 /spl mu/s) high throughput (118.4 MByte/s), performing, cost-effective alternative proprietary high-speed networks, e.g. Myrinet, wide range computing applications.

参考文章(12)
Giuseppe Ciaccio, Messaging on Gigabit Ethernet: Some Experiments with GAMMA and Other Systems Cluster Computing. ,vol. 6, pp. 143- 151 ,(2003) , 10.1023/A:1022804622471
Jonathan Kay, Joseph Pasquale, The importance of non-data touching processing overheads in TCP/IP Computer Communication Review. ,(1993) , 10.1145/167954.166262
P. D. V. van der Stok, M. Boosten, R. W. Dobinson, MESH: MEssageing and ScHeduling for Fine-Grain Parallel Processing on Commodity Plattforms. parallel and distributed processing techniques and applications. pp. 1716- 1722 ,(1999)
Hermann Hellwagner, Markus Lachowitz, Matthias Ohlenroth, Exploring the Performance of VI Architecture Communication Features in the Giganet Cluster LAN. parallel and distributed processing techniques and applications. ,(2000)
Loic Prylli, Bernard Tourancheau, BIP: A new protocol designed for high performance networking on Myrinet international parallel processing symposium. pp. 472- 485 ,(1998) , 10.1007/3-540-64359-1_721
Mario Lauria, Scott Pakin, Andrew Chien, Efficient layering for high speed communication: the MPI over Fast Messages (FM) experience Cluster Computing. ,vol. 2, pp. 107- 116 ,(1999) , 10.1023/A:1019018423211
Shinji Sumimoto, Atsushi Hori, Toshiyuki Takahashi, Yutaka Ishikawa, Hiroshi Harada, PM2: a high performance communication middleware for heterogeneous network environments conference on high performance computing (supercomputing). pp. 16- 16 ,(2000) , 10.5555/370049.370075
Jonathan Kay, Joseph Pasquale, The importance of non-data touching processing overheads in TCP/IP acm special interest group on data communication. ,vol. 23, pp. 259- 268 ,(1993) , 10.1145/166237.166262
Piyush Shivam, Pete Wyckoff, Dhabaleswar Panda, EMP Proceedings of the 2001 ACM/IEEE conference on Supercomputing (CDROM) - Supercomputing '01. pp. 57- 57 ,(2001) , 10.1145/582034.582091
G. Chiola, G. Ciaccio, Efficient parallel processing on low-cost clusters with GAMMA active ports parallel computing. ,vol. 26, pp. 333- 354 ,(2000) , 10.1016/S0167-8191(99)00108-8