Multiple serial jobs
From HP-SEE Wiki
(→High Throughput Computing on BlueGene/P) |
(→High Throughput Computing on BlueGene/P) |
||
Line 1: | Line 1: | ||
=== High Throughput Computing on BlueGene/P === | === High Throughput Computing on BlueGene/P === | ||
- | On IMB BlueGene/P is installed new software called High Throughput Computing (HTC). It enabled portion of the machine to be allocated for many independent single-node jobs. It is used with configuration on compelling scripts and return response for the scripts configuration parameters. | + | On IMB BlueGene/P is installed new software called High Throughput Computing (HTC). It enabled portion of the machine to be allocated for many independent single-node jobs. It is used with configuration on compelling scripts and return response for the scripts configuration parameters. HTC can be used for long-running jobs that require many tasks, this can mean the difference between having to start from scratch and just being able to proceed ahead on the remaining nodes. The front-end node has more memory, better performance, and more functionality than a single compute node. Code that runs on the compute nodes is much cleaner, since it only contains the work to be performed, and leaves the coordination to a script or scheduler. HTC provides 128 computing nodes/CPUs. Basic it allows parallel computing for multiple users. It is possible to make BlueGene/P work like cluster using HTC. It isn’t use for multiple jobs yet. |
=== High Throughput Computing on HPC clusters === | === High Throughput Computing on HPC clusters === |
Revision as of 13:17, 10 April 2012
High Throughput Computing on BlueGene/P
On IMB BlueGene/P is installed new software called High Throughput Computing (HTC). It enabled portion of the machine to be allocated for many independent single-node jobs. It is used with configuration on compelling scripts and return response for the scripts configuration parameters. HTC can be used for long-running jobs that require many tasks, this can mean the difference between having to start from scratch and just being able to proceed ahead on the remaining nodes. The front-end node has more memory, better performance, and more functionality than a single compute node. Code that runs on the compute nodes is much cleaner, since it only contains the work to be performed, and leaves the coordination to a script or scheduler. HTC provides 128 computing nodes/CPUs. Basic it allows parallel computing for multiple users. It is possible to make BlueGene/P work like cluster using HTC. It isn’t use for multiple jobs yet.