Project

General

Profile

Geant4 EM Physics Code Review Weekly Phone Meeting

Meeting

Date: July 26, 2013 at 9.00 AM US CDT (Fermilab local time)
Place: 1-866-740-1260 (ReadyTalk line) (Host : Boyana Norris)

Participants:

Andrea Dotti
Krzysztof Genser
Boyana Norris
Soon Yung Jun

Discussion (Summary by Krzysztof)

Boyana: working with the G4 10beta(9.6.ref06):

Replacing vectors with arrays resulted only in a small (relative)
difference of ~0.5-1.0%

Combining bin/dataVectors (as std::vectors) resulted in ~1.5% global improvement
(using cmsExp g4hpcbenchmark)

The changes involved ~10 files; using macros for now

Will try to optimize spline and log (will move to 9.6.ref07 first though)
Will check in the changes by Monday

Andrea: the local transformations using ?: operator etc... did not
result in a visible time improvement (but we should keep them as the
code is more clear that way)

Krzysztof: will try to take Boyana's changes (Monday) etc... and
prepare the code for a full profiling test (like the ones Soon has
been doing for the reference releases)

Soon: will authorize Boyana and Krzysztof to have write access the
project Redmine site:
https://cdcvs.fnal.gov/redmine/projects/g4hpcbenchmarks

We shall meet again on Friday (10am Central).

(I think we would use the ReadyTalk # provided by Boyana, unless
Boyana objects. Boyana?)