This page should describe Performance Assessment Tests performed for this stand alone feature and should provide links to all the result pages.
Table of Contents |
---|
Summary
Short summary of what was done and what was the resultNew feature tested in MPAS-Ocean stand-alone at low resolution.
Performance
...
Testing
Performance
...
Testing: Timing test in MPAS-Ocean
Date last modified:
Contributors:
...
Provenance: (Run provenance Link, Code Tag, etc:)
Results: (link to results, data and plots)
How was XXX be tested? i.e. how do we know when we have met requirement XXX. Will these unit tests be included in the ongoing going forward?
Performance Test 2
Performance Test 2: short-desciption-of-testing-here
Date last modified:
Contributors: (add your name to this list if it does not appear)
Provenance: (Run provenance Link, Code Tag, etc:)
Results: (link to results, data and plots)
...
Review the number of loops added, verify correct index order. Compare before/after in MPAS-Ocean stand alone with new feature on.
This was done with these timers.
Before:
3 gm bolus velocity 0.03464
After:
3 gm bolus velocity 0.04092
2 time integration 1.86810
Total time increase for this test, which is 6 timesteps in the low-resolution nightly regression suite, is 0.0044/1.868 = 0.0023, or 2.3% of the model integration time for the ocean.
The default performance of GM on compy is here
"o_i:gm bolus velocity" - 320 320 3.200000e+02 5.133580e+00 0.019 ( 1452 0) 0.009 ( 1579 0)
Wallclock total is 5.13358
With the new GM changes
"o_i:gm bolus velocity" - 320 320 3.200000e+02 5.611241e+00 0.025 ( 1486 0) 0.010 ( 1600 0)
Wallclock total is 5.611241
This is a 8.5% increase in the GM timer. However the ocean timestep timer is about 1e7, suggesting these changes will have negligible performance impact.