Updated mdrun -npme documentation

The number of nodes at which PME nodes are used has increased.
Added note on PME nodes not being selected automatically with GPUs.

Fixes #1374.

Change-Id: Ie1de87abd3d1204d99af8b4f8e6809e7806f5c08

diff --git a/src/kernel/mdrun.c b/src/kernel/mdrun.c
index d0645b3bf5f715e56cc71ac6bb9d9e967f8a9bc1..eb30fc99c6c6c360231f0b321772261f79b386c0 100644 (file)
--- a/src/kernel/mdrun.c
+++ b/src/kernel/mdrun.c
@@ -199,13 +199,16 @@ int cmain(int argc, char *argv[])
         "[PAR]",
         "When PME is used with domain decomposition, separate nodes can",
         "be assigned to do only the PME mesh calculation;",
-        "this is computationally more efficient starting at about 12 nodes.",
+        "this is computationally more efficient starting at about 12 nodes",
+        "or even fewer when OpenMP parallelization is used.",
         "The number of PME nodes is set with option [TT]-npme[tt],",
         "this can not be more than half of the nodes.",
         "By default [TT]mdrun[tt] makes a guess for the number of PME",
-        "nodes when the number of nodes is larger than 11 or performance wise",
-        "not compatible with the PME grid x dimension.",
-        "But the user should optimize npme. Performance statistics on this issue",
+        "nodes when the number of nodes is larger than 16. With GPUs,",
+        "PME nodes are not selected automatically, since the optimal setup",
+        "depends very much on the details of the hardware.",
+        "In all cases you might gain performance by optimizing [TT]-npme[tt].",
+        "Performance statistics on this issue",
         "are written at the end of the log file.",
         "For good load balancing at high parallelization, the PME grid x and y",
         "dimensions should be divisible by the number of PME nodes",