There is one active master node, namely the hpcpower2.hku.hk. It is reserved for user login, program modification, compilation and job queue submission/manipulation etc. There is also another passive master node would take up active master node service when there is hardware failure for the active master node.

There are 24 compute nodes (master nodes included) which are used for batch processing, located in two IBM blade centre-E chassis. Except master node, users cannot login to these nodes, but users can submit jobs to be executed in these nodes through the Torque system at master node.

Each compute node is a IBM HS21 blade server that consists of:

  • A IBM HS21 blade server
  • Two 64-bit quad-core Intel Xeon CPU running at 3GHz
  • Two 12MB L2 cache
  • 8GB RAM
  • Two 146 GB SAS hard disk
  • Dual integrated 10/100/1000 ethernet ports

File Service:

  • A 4TB iSCSI storage of IBM N3600 storage appliance
  • A pair of dedicated HA NFS file servers of IBM x3550 servers

Network:

  • Two groups of gigabit ethernet networks are formed, one is for the private NFS file system and one is for the inter-processor MPI communication.

Directories:

  • /home1 and /home2 are used for user home directories and files.
  • /share1 and /share2 are used to store commonly used files.
  • /sharedtmp1 and /sharetmp2 can be read and write from all nodes. Its size is 28GB.
  • Each slave node has 30GB of local /tmp directory.