The HKU Gideon 300 Cluster 

- A Self-made Commodity Supercomputer


Visit more updated information about Gideon 300 :







May 2, 2003 Rod Greenwood's visit : head of Morgan Stanley IT Enterprise Infrastructure.
April 11, 2003 Hong Kong Linux Industry Association (HKLIA) visited HKU SRG (photo)
March 11, 2003 Morgan Stanley (HK): HKU grid computing/supercomputer field trip

1. Introduction

Gideon 300 cluster is a high-performance cluster computer, which is built by the Systems Research Group (SRG) of the Department of Computer Science and Information Systems at The University of Hong Kong. The construction of Gideon started in early-October and became fully operational in mid-October. Gideon 300 will be used to conduct research at HKU on new classes problems that are only solvable using a high degree of parallelism and "supercomputing" power.

2. What is a cluster ?

A cluster is a type of parallel or distributed processing system, which consists of a collection of interconnected stand-alone computers working together as a single, integrated computing resource.

Words from Dr. Francis Lau (Head of CSIS)--- Clusters of today are an important element of mainstream computing. In less than a decade, the "lowly" computer clusters have come to prominence as the most convenient and cost-effective tool for solving complex computational problems. Being able to bring together enough computing cycles together in a pool is one thing, to be able to use these cycles efficiently so that the aggregate power of a cluster can be fully harnessed is another. Research and development effort must continue to explore better strategies to make clusters an even greater tool when facing up to the grandest challenges of this age.

3. Gideon 300 Cluster

The cluster consists of 300 Intel-based Legend PCs interconnected by a single, high-port density Fast Ethernet switch. This guarantees a one-hop latency between each node with full connectivity. Each processing node is merely a standard PC which has one Intel Pentium IV processors running at 2 Ghz, 512 Mbytes DDR memory, 40GB IDE harddisk. This gives the whole cluster a peak floating point performance of 1.2 Tflops (trillion floating point operations per second). The cluster has been benchmarked according to an international standard - the High-Performance Linpack benchmark. This benchmark test is used by the Top500 list ( to rank the world's 500 fastest computers according to their benchmark performance.  The Gideon 300 cluster is now ranked 175 on the latest TOP500 Supercomputers  list (11/2002). We believe it is the fastest PC cluster in the academic field in Hong Kong.

4. Technical Data

Node Configuration :

Network Components

File Servers

Performance Data

5. Ongoing Research Projects at Gideon 300 Cluster

6. Building the Gideon Cluster

If you are interested in knowing how we build the Gideon cluster during the beginning stage, please visit this page.

7. Photo Collection

Oct. 9, 2002 


Oct. 11, 2002

OS Installation

Oct. 17, 2002

Linpack Test (256-node : 306 Gflops)

Oct. 19, 2002

HKU Open Day

8. About the name of the cluster

We call the 300-node cluster "HKU Gideon Cluster" or "HKU Gideon300 Cluster". The name Gideon 300 is derived from Gideon's 300 warriors.

9. Further information

See also the SRG web page ( for some more information about the new Gideon 300 cluster. If you have any questions about the new cluster please send a mail to: Dr. Cho-Li Wang ( or Dr. Anthony Tam (