Abstract. High speed networks are now providing incredible performances. Software evolution is slow and the old protocol stacks are no longer adequate for these kind of communication speed. When bandwidth increases, the latency should decrease as much in order to keep the system balance. With the current network technology, the main bottleneck is most of the time the software that makes the interface between the hardware and the user. We designed and implemented new protocols of transmission targeted to parallel computing that squeeze the most out of the high speed Myrinet network, without wasting time in system calls or memory copies, giving all the speed to the applications. This design is presented here as well as experimental results that lead to achieve real Gigabit/s throughput and less than 5 s latency on a cluster of PC workstations, with this a ordable network hardware. Moreover, our networking results compare favorably with the expensive parallel computers or ATM LANs. 1

High speed networks are now providing incredible performance. Software evolution is slow and the old protocol stacks are no longer adequate for these kind of communication speeds. When bandwidth increases, the latency should decrease as much in order to keep the system balance. With the current network technology, the main bottleneck is most often the software that is the interface between the hardware and the user. We designed and implemented new transmission protocols, targeted to parallel computing, that squeeze the most out of a high speed network (Myrinet in this paper) without wasting time in system calls or memory copies, giving all the speed to the applications. This design is presented here as well as experimental results. We achieve Gigabit/s bandwidth and less than 5s latency on a cluster of PC workstations with inexpensive network hardware. Moreover, our results compare favorably with the expensive parallel computers or ATM LANs. Keywords: High-speed networks, Co...