BrightComputing has announced that the University of Houston’s Texas Learning and Computation Center [TLC2] has chosen their Bright Cluster Manager as the preferred management stack for their clustered computing systems. The Cluster Manager stack was installed on a 160 node cluster at TLC2 in under a day over the summer.
I am very impressed with the efficiency achieved with Bright Cluster Manager. Our cluster with 1280 cores was up and running within a few hours, ready for integration into our HPC environment. Now it is continuing to save our system administrators valuable time,” said Professor Lennart Johnsson, Director of the TLC2 and the Advanced Computing Research Laboratory at the University of Houston.
The new machine at TLC2, called Xanadu, consists of 160 Dell PowerEdge servers with AMD Opteron Barcelona-flavored CPUs.
We have tried many different cluster management solutions in the past. We found Bright Cluster Manager to have a well integrated centralized management interface which is powerful and flexible enough to accommodate our specific configuration requirements. For example, we like to setup our clusters with specific network configurations. Unlike previously used solutions, with Bright Cluster Manager this was very easy to configure through the intuitive GUI or the powerful command line shell.”
Cluster management has been very hit or miss throughout the Beowulf history. Those that perform the task well, do so very well. Those that don’t, well, don’t. For more info, read the full release here at HPCWire.