High-performance ethernet-based communications for future multi-core processors

作者: Michael Schlansker , Norman P. Jouppi , Nagabhushan Chitlur , Erwin Oertli , Paul M. Stillwell

DOI: 10.1145/1362622.1362672

关键词: Carrier EthernetComputer networkEmbedded systemIndustrial EthernetMetro EthernetEthernet over SDHEthernet over PDHATA over EthernetComputer scienceConnection-oriented EthernetSynchronous Ethernet

摘要: Data centers and HPC clusters often incorporate specialized networking fabrics to satisfy system requirements. However, Ethernet's low cost high performance are causing a shift from toward standard Ethernet. Although low-level approaches that of fabrics, the features these provide such as reliable in-order delivery flow control implemented, in case Ethernet, by endpoint hardware software. Unfortunately, current Ethernet endpoints either slow (commodity NICs with generic TCP/IP stacks) or costly (offload engines). To address issues, JNIC project developed novel endpoint. JNIC's software were specifically designed for requirements high-performance communications within future data-centers compute clusters. The architecture combines capabilities already seen advanced network architectures new innovations create comprehensive solution scalable We envision is suitable most in-data-center communication needs.

参考文章(19)
P. Balaji, W. Feng, Q. Gao, R. Noronha, W. Yu, D. K. Panda, Head-to-TOE Evaluation of High-Performance Sockets over Protocol Offload Engines international conference on cluster computing. pp. 1- 10 ,(2005) , 10.1109/CLUSTR.2005.347068
Cyriel Minkenberg, Mitchell Gusat, None, Speculative Flow Control for High-Radix Datacenter Interconnect Routers international parallel and distributed processing symposium. pp. 1- 10 ,(2007) , 10.1109/IPDPS.2007.370260
L. Cherkasova, V. Kotov, T. Rokicki, Fibre channel fabrics: evaluation and design hawaii international conference on system sciences. ,vol. 1, pp. 53- 62 ,(1996) , 10.1109/HICSS.1996.495447
Shubhendu S. Mukherjee, Babak Falsafi, Mark D. Hill, David A. Wood, Coherent Network Interfaces for Fine-Grain Communication international symposium on computer architecture. ,vol. 24, pp. 247- 258 ,(1996) , 10.1145/232973.232999
Jiuxing Liu, B. Chandrasekaran, Weikuan Yu, Jiesheng Wu, D. Buntinas, S. Kini, D.K. Panda, P. Wyckoff, Microbenchmark performance comparison of high-speed cluster interconnects IEEE Micro. ,vol. 24, pp. 42- 51 ,(2004) , 10.1109/MM.2004.1268994
In-Su Yoon, Sang-Hwa Chung, Ben Lee, Hyuk-Chul Kwon, Implementation and Performance Evaluation of M-VIA on AceNIC Gigabit Ethernet Card european conference on parallel processing. pp. 995- 1000 ,(2003) , 10.1007/978-3-540-45209-6_136
Patricia Gilfeather, Arthur B. Maccabe, Making TCP Viable as a High Performance Computing Protocol ,(2002)
Jack Dongarra, Steven Huss-Lederman, David W. Walker, Steve W. Otto, Marc Snir, MPI: The Complete Reference ,(1996)
E. Gronke, G. McAlpine, A.M. Merritt, B. Shubert, F. Berry, C. Dodd, D. Dunning, G. Regnier, D. Cameron, The virtual interface architecture ,(2002)
Sejin Park, Sang-Hwa Chung, Ben Lee, Implementation and performance study of a hardware-VIA-based network adapter on gigabit ethernet Journal of Systems Architecture. ,vol. 51, pp. 602- 616 ,(2005) , 10.1016/J.SYSARC.2005.02.002