Present: Vassili, Peter, Erik, Philipp, Yosef, Roland new Carpet version:
* Philipp observed increased memory consumption (run terminated by Out-Of-Memory killer) and a run speed that was continuously decreasing over the time of the run on bluewaters. This used the Intel compiler and ~8000 cores * Roland looked at differences in data between "old" and "new" versions, sees roundoff level differences in rho (for a NsNs run) but larger ones in eps. Could just be the usual case of eps being more sensitive. Roundoff level differences are expected due to different processor decomposition New thorns: * Erik proposed a MemSpeed thorn on trac * Yosef's ParitySymmetry thorn is still up for review * please review the changes to Regrid2 for ParitySymmetry Calls: * very little to discuss these days, consider having the meeting only every two weeks? Please comment. Yours, Roland
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Users mailing list [email protected] http://lists.einsteintoolkit.org/mailman/listinfo/users
