Articles | Volume 8, issue 3
Model experiment description paper
24 Mar 2015
Model experiment description paper |  | 24 Mar 2015

Efficient performance of the Met Office Unified Model v8.2 on Intel Xeon partially used nodes

I. Bermous and P. Steinle


Interactive discussion

Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision
AR by Ilia Bermous on behalf of the Authors (28 Jan 2015)  Author's response    Manuscript
ED: Referee Nomination & Report Request started (10 Feb 2015) by Dan Lunt
RR by Anonymous Referee #2 (20 Feb 2015)
ED: Publish subject to minor revisions (Editor review) (24 Feb 2015) by Dan Lunt
AR by Ilia Bermous on behalf of the Authors (05 Mar 2015)  Author's response    Manuscript
ED: Publish as is (05 Mar 2015) by Dan Lunt
Short summary
The trend in High Performance Computing (HPC) is for less memory bandwidth relative to the computational power of each core. With each CPU having multiple cores, the best way of using HPC systems is not always straightforward. For some time critical applications, shorter run times can be obtained by using only some of the cores per CPU and keeping the others idle. A number of factors are required to consider, but this provides a simple technique for a significant gain in the application speed.