Does Gromacs use the Phi coprocessor on stampede? Chris Neale https://conferences.xsede.org/c/message_boards/find_message?p_l_id=&messageId=500679 2013-04-17T17:25:55Z 2013-04-17T17:25:55Z

Hi,<br /><br />I just tried using gromacs 4.5.5 on stampede for the first time (using the supplied module). I ran like this:<br /><br />ibrun mdrun_mpi -notunepme -deffnm md3 -nosum -dlb yes -npme -1 -cpt 60 -maxh 0.1 -cpi md3.cpt -nsteps 5000000000<br /><br />From looking at the speed (ns/day) I get approx. what I would expect from 16 cores alone, so I suspect that the MICs are not being used. I also looked at the .log information from gromacs (see below), which also doesn&#039;t list the MICs. Does anybody know if they are being used, or how I can check that?<br /><br />Thank you,<br />Chris.<br /><br /> R E A L C Y C L E A N D T I M E A C C O U N T I N G<br /><br /> Computing: Nodes Number G-Cycles Seconds %<br />-----------------------------------------------------------------------<br /> Domain decomp. 12 2281 304.668 112.8 2.0<br /> DD comm. load 12 2280 1.257 0.5 0.0<br /> DD comm. bounds 12 2282 2.724 1.0 0.0<br /> Send X to PME 12 22801 31.611 11.7 0.2<br /> Comm. coord. 12 22801 75.950 28.1 0.5<br /> Neighbor search 12 2281 1199.547 444.3 7.8<br /> Force 12 22801 6957.146 2576.8 45.1<br /> Wait + Comm. F 12 22801 807.014 298.9 5.2<br /> PME mesh 4 22801 1159.146 429.3 7.5<br /> Wait + Comm. X/F 4 2693.618 997.7 17.5<br /> Wait + Recv. PME F 12 22801 18.672 6.9 0.1<br /> Write traj. 12 2 0.321 0.1 0.0<br /> Update 12 22801 320.949 118.9 2.1<br /> Constraints 12 45602 1532.245 567.5 9.9<br /> Comm. energies 12 2282 4.756 1.8 0.0<br /> Rest 12 301.601 111.7 2.0<br />-----------------------------------------------------------------------<br /> Total 16 15411.226 5708.0 100.0<br />-----------------------------------------------------------------------<br />-----------------------------------------------------------------------<br /> PME redist. X/F 4 45602 113.086 41.9 0.7<br /> PME spread/gather 4 45602 640.632 237.3 4.2<br /> PME 3D-FFT 4 45602 306.614 113.6 2.0<br /> PME solve 4 22801 98.323 36.4 0.6<br />-----------------------------------------------------------------------<br /><br /> Parallel run - timing based on wallclock.<br /><br /> NODE (s) Real (s) (%)<br /> Time: 356.752 356.752 100.0<br /> 5:56<br /> (Mnbf/s) (GFlops) (ns/day) (hour/ns)<br />Performance: 714.533 44.446 11.044 2.173

Chris Neale 2013-04-17T17:25:55Z