User Tools

Site Tools


wg:geodynamo:performance_results

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
wg:geodynamo:performance_results [2014/10/27 16:51]
kellogg
— (current)
Line 1: Line 1:
-These pages show the results of performance tests for different dynamo codes on the [[https://​www.tacc.utexas.edu/​user-services/​user-guides/​stampede-user-guide|TACC XSEDE Stampede system]]. 
  
-To clarify terminology,​ here are some definitions of terms used in the measurements:​ 
-  * Node: On Stampede, each node contains //2 Xeon E5-2680// processors, each with 8 cores, for a total of 16 cores per node.  Each node also has //32GB 8x4G 4 channels DDR3-1600MHz//​ of memory, ​ 
-  * Core: A single core in a processor. 
-  * MPI processes: ​ 
-  * Thread: An OpenMP thread. Unless otherwise specified processor thread affinity is not enabled. 
-  * SU: Service unit, a unit of usage accounting equal to //#cores x #hours//. Usage is billed by the node, so using 8 of 16 cores in a node for 1 hour will be recorded as 16 SUs rather than 8 SUs. 
- 
-All timings are measured in terms of wall time unless otherwise specified. They are measured using ''​MPI_Wtime()''​ (If program does not use MPI, ''​system_clock(itime,​ irate, imax)''​ is used). The time for one step is determined by taking the mean of the elapsed time for 100 time steps. 
- 
-[[wg:​geodynamo:​Performance_results:​Calypso|Calypso Ver.1.1]] \\ 
-Status: Three cases should be tested again \\ 
- 
-[[wg:​geodynamo:​Performance_results:​Calypso_Latest|Calypso latest Ver.]] \\ 
-Status: Two cases are missing and three cases should be tested again \\ 
- 
- 
-[[wg:​geodynamo:​Performance_results:​GeoFEM|GeoFEM-MHD]] \\ 
-Status: Done \\ 
- 
-[[wg:​geodynamo:​Performance_results:​ETH|ETH code (not finished)]] \\ 
-Status: Having a problem to generate initial field. Asking to authors. \\ 
- 
-[[wg:​geodynamo:​Performance_results:​Dennou|SPmodel(GFD-dennou code) (not finished)]] \\ 
-Status: Strong scaling is done. Discussing with authors \\ 
- 
-[[wg:​geodynamo:​Performance_results:​Goddard|Goddard code]] \\ 
-Status: Routine has been written to generate initial field of multiple resolutions. ​ 
-Tests are underway. ​ 
-Request pending for how to deal with dt adjustment. ​ 
- 
-[[wg:​geodynamo:​Performance_results:​H2000|H2000]] \\ 
-Status: Done \\ 
- 
-[[wg:​geodynamo:​Performance_results:​IPGP|Parody(IPGP JHU code)]] \\ 
-Status: Done \\ 
- 
-[[wg:​geodynamo:​Performance_results:​LSD|LSD (Leeds code)]] \\ 
-Status: Sending the report to authors \\ 
- 
-[[wg:​geodynamo:​Performance_results:​MagIC|MagIC]] \\ 
-Status: Done \\ 
- 
-[[wg:​geodynamo:​Performance_results:​MagIC5|MagIC5 (not started)]] \\ 
- 
-[[wg:​geodynamo:​Performance_results:​SFEMANS|SFEMANS (not finished)]] \\ 
-Status: Strong scaling is done. Discussing with authors for warnings and changing resolutions \\ 
- 
-[[wg:​geodynamo:​Performance_results:​Busse|simitev-busse-drs]] \\ 
-Status: Strong scaling is done with $l_{max} = 128$. Simulations fail more than $l_{max} = 192$. 
- 
-[[wg:​geodynamo:​Performance_results:​Gary|UCSC-Glatzmaier code]] \\ 
-Status: Strong scaling is done. 
- 
-[[wg:​geodynamo:​Performance_results:​Gary_Latest|UCSC-Glatzmaier code (New Version)]] \\ 
-Status: Code received. 
- 
-[[wg:​geodynamo:​Performance_results:​xshells|xshells]] \\ 
-Status: Done, need to communicate with authors. \\ 
- 
-[[wg:​geodynamo:​Performance_results:​TITECH|TITECH code (not started)]] \\ 
- 
-[[wg:​geodynamo:​Performance_results:​Comp|Comparison among codes]] \\ 
wg/geodynamo/performance_results.1414428697.txt.gz ยท Last modified: 2014/10/27 16:51 by kellogg