- 27 Jan, 2016 5 commits
-
-
Matthieu Schaller authored
Update single node task plots Update the non-MPI task plot scripts to offer same speed ups as the MPI versions. See merge request !83
-
Peter W. Draper authored
-
Peter W. Draper authored
-
Peter W. Draper authored
Fix MPI tag collision Issue #68 Increased the maximum cell tag to 2^29 to prevent the MPI tag collision issue #68. See merge request !82
-
James Willis authored
-
- 22 Jan, 2016 17 commits
-
-
Matthieu Schaller authored
PASC paper See merge request !80
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Matthieu Schaller authored
-
Pedro Gonnet authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Peter W. Draper authored
Corrected the Wendland C2 kernel. Still needs checking. @jwillis, can you check whether the number of neighbours is sensible with this version ? The definition of kernels is still not completely optimal so minor changes may appear at a later stage in the project when nothing more urgent needs attention. See merge request !77
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
- 21 Jan, 2016 18 commits
-
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Pedro Gonnet authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Peter W. Draper authored
NUMA-aware affinity Resolves #76. Yes, ```hyperthreads_present``` is bonkers - it's easy to detect whether the processor supports the technology, but this is the best I've found to tell whether it's enabled. See merge request !78
-
Angus Lepper authored
* Alternative: does .../thread_siblings contain more than a single 1? n.b. May or may not contain a comma if so.
-
Angus Lepper authored
* This is not vectorisation-specific. * There may be better trade-offs between HT and NUMA. * Maybe print a warning when we require multiple NUMA nodes. * Must also detect when Hyper-Threading is not present. * Probably better as a configure flag, rather than conditional only upon the availability of libnuma. * ~15-40% performance improvement on COSMA.
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-
Matthieu Schaller authored
-