Fast Barrier Synchronization on Shared Fast Ethernet

摘要

Shared LAN is presently the most widespread networking technology, due to its extremely low cost and favourable cost/performance ratio. Clusters of personal Computers leveraging shared 100Base-T Ethernet may currently offer the best price/performance in parallel processing. Most numberical parallel algorithms make heavy use of collective ocmmunications and especially barrier synchronization. Hence a critical issue on PC clusters is to offer efficient implementations of such primitives even though using low-cost, non-switched LAN technology. We implemented and studied some simple barrier synchronization protocols atop the Genoa Active Message MAchien, an efficient Active Messges-like communication layer running on a cluster of pentium PCs connected by a 100base-TX Ethernet repeater hub. In the case of synchronized or quasi-synchronized processes issuing a barrier synchronization, an obvious way to avoid collisiions on shared 100base-T Ethernet is to use a barrier protocol which explicitly serializes all the intr-process synchronziation commuications over the LAN. We propose alternative barrier protocols which avoid Ethernet collisions during the synchronization phase without requiring such a ful explicit serialization. One of such protocols definitely outperforms the fully explicit serialization. One of such protocols definitely outperforms the fully serialized barrier protocol over 100base-T Ethernet as well as the MIP implementation of barrier synchronization on IBM SP2 and Intel paragon.