The PACS-CS Project

TL;DR

The paper outlines the development of a large-scale, high-performance cluster system for lattice QCD simulations, significantly surpassing previous computational capabilities at the University of Tsukuba.

Contribution

It presents a detailed plan for a 14.3 Tflops cluster system with innovative network architecture and hardware specifications for advanced scientific computing.

Findings

Design of a 2560-node cluster with high bandwidth network

Implementation of Linux and SCore for system operation

Scheduled deployment in 2006

Abstract

We describe our plan to develop a large-scale cluster system with a peak speed of 14.3Tflops for lattice QCD at the Center for Computational Sciences, University of Tsukuba, as a successor to the current 0.6Tflops CP-PACS computer. The system consist of 2560 nodes connected by a 16x16x10 three-dimensional hyper crossbar network. Each node has a single low-voltage 2.8GHz Xeon processor and 2GBytes of memory with 6.4GBytes/sec bandwidth, and 160 GBytes of disk in RAID1 mode. The network link in each of the three directions is made of dual Gigabit Ethernet with the peak throughput of 250MByte/sec. Hence each node has an aggregate network bandwidth of 750MByte/sec. The system will run under Linux and SCore, and an extension of the PM driver is developed for the network. The system will be developed jointly with Hitachi Limited. The installation is scheduled in the first quarter of Japanese Fiscal 2006 (April-June 2006) and the start of operation is expected in July 2006.