Design and Implementation of a Synchronization and Communication Controller for Cluster Computing Systems

摘要

This paper describes a new network design for low-latency synchronization and communication and descirbes the mechanism of synchronization on a Synchroniza-tion/Communication Controller (SCC) that can be plugged into each workstation/PC connected via conventional LAN. Each SCC is connected by a new type of network (sync-conn network). Each node sends high-throughput data to the desired destination over a conventional LAN us-ing TCP/IP, and synchronizes and sends low-latency data via the sync-comm network. SCC requires only 630 ns to perform barrier synchronization with 16 nodes. In case of including software overhead, when SCC boards were plugged into each PC in a 4-nodes cluster, the barrier la-tency was only 6.2mus. Since the SCC is a PCI board, it can be quickly plugged into almost any workstation/PC cluster with a PCI bus slot.