ETHERNET STATUS SAYS I HAVE 1GBPS. ONLY GETTING 1.8 SOFTWARE
Hiroshi Tezuka: He is a Senior Researcher of Parallel and Distributed System Software Laboratory at Real World Computing Partnership, JAPAN. He is a member of Information Processing Society of Japan. His research interest include parallel and distributed systems, real-time systems, and high performance communication facilities.
He received BS degree in electrical engineering from Doshisha University. Shinji Sumimoto: He is a Senior Researcher of Parallel and Distributed System Software Laboratory at Real World Computing Partnership, JAPAN. and Li, K., “VMMC-2: Efficient Support for Reliable, Connection-Oriented Communication,” Hot Interconnect’97, August 1997. and Tourancheau, B., “BIP: a new protocol designed for high performance,” PC-NOW Workshop, help in parallel with IPPS/SPDP98, Orlando, USA, March 30–April 3 1998.ĭubnicki C., Bilas, A., Chen, Y., Damianakis, S. E., “Virtual Network Transport Protocols for Myrinet,” Hot Interconnect’97, August 1997. of the Third International Symposium on High Performance Computer Architecture (HPCA), February 1997.Ĭhun, B.
von, “U-Net: A User-Level Network Interface for Parallel and Distributed Computing,” Proc. and Ishikawa, Y., “Pin-down Cache: A Virtual Memory Management Technique for Zero-copy Communication,” IPPS/SPDP’98, IEEE, pp. (ed.), High-Performance Computing and Networking, 1225, Lecture Notes in Computer Science, pp.
and Sato, M., “PM: An Operating System Coordinated High Performance Communication Library,” Sloot, P. of Supercomputing ’95, San Diego, California, 1995. and Chein, A., “High Performance Messaging on Workstations: Illinois Fast Messages (FM) for Myrinet,” Proc. of Hot Interconnects II, 1994 Palo Alto, August 1994. and Buch, V., “Low-Latency Communication over ATM Networks Using Active Messages,” Proc. E., “Active Messages: A Mechanism For Integrated Communication And Computation,” Proc. and Ishikawa, Y., “The Design and Evaluation of High Performance Communication Using a Gigabit Ethernet,” International Conference on Supercomputing ’99, pp. Sumimoto, S., Tezuka, H., Hori, A., Harada, H., Takahashi, T. The results show that the IS class S performance on the GigaE PM is 1.8 times faster than that on TCP/IP. We have implemented MPICH-PM on top of the GigaE PM, and evaluated the NAS parallel benchmark performance. The performance results show that a 58.3 μs round trip time for a four byte user message, and 56.7 MBytes/sec bandwidth for a 1,468 byte message have been achieved on Intel Pentium II 400 MHz PCs. A prototype system has been implemented using an Essential Communications Gigabit Ethernet card. A reliable communication mechanism for a parallel application is implemented on the firmware on a NIC while existing network protocols are handled by an operating system kernel. The GigaE PM provides not only a reliable high bandwidth and low latency communication, but also supports existing network protocols such as TCP/IP. A high performance communication facility, called the GigaE PM, has been designed and implemented for parallel applications on clusters of computers using a Gigabit Ethernet.
0 kommentar(er)
