Introduction
An IBM eserver 1350 cluster is available for compute-intensive simulations and data analysis tasks. The cluster, named Kronos running under Linx operation system, consists of 28 computing nodes each with 8 Xeon 2.3 GHZ cores and 12 GB of memory. Additionally, two nodes, each with 8 Xeon 2.5 GHZ cores and 64 GB memory are available for memory intensive tasks. These nodes are interconnected by Gigabit Ethernet and feature a 35 TB Global Parallel File System (GPFS). This system provides an aggregate 240 CPUs and 464 GB of memory.
These systems are configured to provide a development and job running environment similar to those that are available in other HPC centers. Portland Group, Intel and GNU Fortran, C/C++ compilers, numerical libraries including Portland groups ACML, Intel's Math Kernel Library (MKL), GSL and IO libraries including NetCDF and HDF are installed on these systems. The MOAB job scheduler and resource management system is used manage production jobs.
The most up-to-date information on the current status of Kronos can be viewed
here.
Figure 1 presents an example on the status of Kronos at on 16 July 2009,
and the system is under heavy usage. Note that the system status can only be
accessed within the university network. If you want to obtain the system status from an
off-campus computer; e.g., at home or during the travel, you will have to set up a VPN
connection to the university first. More information about
the university VPN connection can be found
here.
Free free to contact us to report any technical difficulties
or abnormal patterns you happen to notice from the system status.
|
|
