Difference between revisions of "Namd on BGQ"
Line 1: | Line 1: | ||
− | Taking into consideration the official (test)[NAMD performance tuning recommendations], a parameter study was undertaken to test simulation performance and efficiency on the BG/Q cluster. | + | Taking into consideration the official (test)[NAMD performance tuning recommendations], a parameter study was undertaken to test simulation performance and efficiency on the BG/Q cluster. Efficiency was measured with respect to the 16 ranks-per-node 512 core simulation. All simulations were conducted from a restart trajectory from an equilibrated system. |
− | The following study was undertaken for a 246,000 atom membrane protein (Cytochrome c | + | The following study was undertaken for a 246,000 atom membrane protein simulation ([http://www.rcsb.org/pdb/explore.do?structureId=1m56 Cytochrome c Oxidase] embedded in a TIP3P solvated DPPC bilayer) using the CHARMM36 forcefield (protein and lipids). The unit cell is cubic with box dimensions 144 x 144 x 117 Angstroms. Long-range electrostatics are computed using PME for all simulations with PME grid spacing set to be generated automatically with the '''pmeGridSpacing 1.0''' setting. A poor choice in PME grid spacing (i.e. not a multiple of 2,3, and 5) can result in performance degradation of |
{| class="wikitable" | {| class="wikitable" | ||
Line 13: | Line 13: | ||
|512 | |512 | ||
| | | | ||
− | |2. | + | |2.79 |
|1.00 | |1.00 | ||
|---- | |---- | ||
Line 19: | Line 19: | ||
|1024 | |1024 | ||
| | | | ||
− | |5. | + | |5.05 |
|0.91 | |0.91 | ||
|---- | |---- | ||
Line 25: | Line 25: | ||
|1024 | |1024 | ||
|twoAwayX (default) | |twoAwayX (default) | ||
− | |5. | + | |5.62 |
|1.01 | |1.01 | ||
|---- | |---- | ||
Line 31: | Line 31: | ||
|2048 | |2048 | ||
|twoAwayX (default) | |twoAwayX (default) | ||
− | |10. | + | |10.07 |
|0.90 | |0.90 | ||
|---- | |---- | ||
Line 37: | Line 37: | ||
|2048 | |2048 | ||
|twoAwayXY | |twoAwayXY | ||
− | |10. | + | |10.59 |
|0.95 | |0.95 | ||
|---- | |---- | ||
Line 43: | Line 43: | ||
|4096 | |4096 | ||
|twoAwayX | |twoAwayX | ||
− | |14. | + | |14.32 |
|0.64 | |0.64 | ||
|---- | |---- | ||
Line 49: | Line 49: | ||
|4096 | |4096 | ||
|twoAwayXY (default) | |twoAwayXY (default) | ||
− | |17. | + | |17.63 |
|0.79 | |0.79 | ||
|---- | |---- | ||
Line 55: | Line 55: | ||
|4096 | |4096 | ||
|twoAwayXYZ | |twoAwayXYZ | ||
− | |16. | + | |16.79 |
|0.75 | |0.75 | ||
|---- | |---- | ||
Line 61: | Line 61: | ||
|8192 | |8192 | ||
|twoAwayX | |twoAwayX | ||
− | |23. | + | |23.52 |
|0.53 | |0.53 | ||
|---- | |---- | ||
Line 67: | Line 67: | ||
|8192 | |8192 | ||
|twoAwayXY (default) | |twoAwayXY (default) | ||
− | | | + | |25.00 |
|0.56 | |0.56 | ||
|---- | |---- | ||
Line 73: | Line 73: | ||
|16384 | |16384 | ||
|twoAwayX | |twoAwayX | ||
− | |23. | + | |23.67 |
|0.27 | |0.27 | ||
|---- | |---- | ||
Line 79: | Line 79: | ||
|16384 | |16384 | ||
|twoAwayXY | |twoAwayXY | ||
− | |28. | + | |28.31 |
|0.32 | |0.32 | ||
|---- | |---- | ||
Line 85: | Line 85: | ||
|16384 | |16384 | ||
|twoAwayXYZ (default) | |twoAwayXYZ (default) | ||
− | |27. | + | |27.98 |
|0.31 | |0.31 | ||
|---- | |---- | ||
Line 103: | Line 103: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=8, lblUnload=yes | |twoAwayXY, PMEPencils=8, lblUnload=yes | ||
− | |12. | + | |12.93 |
|0.58 | |0.58 | ||
|---- | |---- | ||
Line 109: | Line 109: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=12, lblUnload=yes | |twoAwayXY, PMEPencils=12, lblUnload=yes | ||
− | |17. | + | |17.27 |
|0.77 | |0.77 | ||
|---- | |---- | ||
Line 115: | Line 115: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=16, lblUnload=yes | |twoAwayXY, PMEPencils=16, lblUnload=yes | ||
− | |16. | + | |16.02 |
|0.72 | |0.72 | ||
|---- | |---- | ||
Line 121: | Line 121: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=20, lblUnload=yes | |twoAwayXY, PMEPencils=20, lblUnload=yes | ||
− | |15. | + | |15.41 |
|0.69 | |0.69 | ||
|---- | |---- | ||
Line 139: | Line 139: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=12 | |twoAwayXY, PMEPencils=12 | ||
− | |16. | + | |16.21 |
|0.73 | |0.73 | ||
|---- | |---- | ||
Line 145: | Line 145: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=16 | |twoAwayXY, PMEPencils=16 | ||
− | |17. | + | |17.92 |
|0.80 | |0.80 | ||
|---- | |---- | ||
Line 151: | Line 151: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=20 | |twoAwayXY, PMEPencils=20 | ||
− | |17. | + | |17.99 |
|0.81 | |0.81 | ||
|---- | |---- | ||
Line 157: | Line 157: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=24 | |twoAwayXY, PMEPencils=24 | ||
− | |17. | + | |17.83 |
|0.80 | |0.80 | ||
|---- | |---- | ||
Line 163: | Line 163: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=36 | |twoAwayXY, PMEPencils=36 | ||
− | |16. | + | |16.97 |
|0.76 | |0.76 | ||
|---- | |---- | ||
Line 181: | Line 181: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=20 | |twoAwayXY, PMEPencils=20 | ||
− | |18. | + | |18.24 |
|0.82 | |0.82 | ||
|---- | |---- | ||
Line 187: | Line 187: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=20 | |twoAwayXY, PMEPencils=20 | ||
− | |17. | + | |17.99 |
|0.81 | |0.81 | ||
|---- | |---- | ||
Line 193: | Line 193: | ||
|4096 | |4096 | ||
|twoAwayXY, PMEPencils=20 | |twoAwayXY, PMEPencils=20 | ||
− | |13. | + | |13.94 |
|0.63 | |0.63 | ||
|---- | |---- | ||
Line 205: | Line 205: | ||
|512 | |512 | ||
| | | | ||
− | |2. | + | |2.86 |
|1.03 | |1.03 | ||
|---- | |---- | ||
Line 211: | Line 211: | ||
|512 | |512 | ||
| | | | ||
− | |2. | + | |2.84 |
|1.02 | |1.02 | ||
|---- | |---- | ||
Line 217: | Line 217: | ||
|512 | |512 | ||
| | | | ||
− | |2. | + | |2.79 |
|1.00 | |1.00 | ||
|---- | |---- | ||
Line 223: | Line 223: | ||
|512 | |512 | ||
| | | | ||
− | |2. | + | |2.29 |
|0.82 | |0.82 | ||
|---- | |---- | ||
Line 234: | Line 234: | ||
|16 | |16 | ||
|512 | |512 | ||
− | | | + | |ldbUnloadZero=yes |
− | |2. | + | |2.79 |
|1.00 | |1.00 | ||
|---- | |---- |
Revision as of 17:38, 29 October 2012
Taking into consideration the official (test)[NAMD performance tuning recommendations], a parameter study was undertaken to test simulation performance and efficiency on the BG/Q cluster. Efficiency was measured with respect to the 16 ranks-per-node 512 core simulation. All simulations were conducted from a restart trajectory from an equilibrated system.
The following study was undertaken for a 246,000 atom membrane protein simulation (Cytochrome c Oxidase embedded in a TIP3P solvated DPPC bilayer) using the CHARMM36 forcefield (protein and lipids). The unit cell is cubic with box dimensions 144 x 144 x 117 Angstroms. Long-range electrostatics are computed using PME for all simulations with PME grid spacing set to be generated automatically with the pmeGridSpacing 1.0 setting. A poor choice in PME grid spacing (i.e. not a multiple of 2,3, and 5) can result in performance degradation of
Ranks | Cores | NAMD Config Options | ns/day | Efficiency |
16 | 512 | 2.79 | 1.00 | |
16 | 1024 | 5.05 | 0.91 | |
16 | 1024 | twoAwayX (default) | 5.62 | 1.01 |
16 | 2048 | twoAwayX (default) | 10.07 | 0.90 |
16 | 2048 | twoAwayXY | 10.59 | 0.95 |
16 | 4096 | twoAwayX | 14.32 | 0.64 |
16 | 4096 | twoAwayXY (default) | 17.63 | 0.79 |
16 | 4096 | twoAwayXYZ | 16.79 | 0.75 |
16 | 8192 | twoAwayX | 23.52 | 0.53 |
16 | 8192 | twoAwayXY (default) | 25.00 | 0.56 |
16 | 16384 | twoAwayX | 23.67 | 0.27 |
16 | 16384 | twoAwayXY | 28.31 | 0.32 |
16 | 16384 | twoAwayXYZ (default) | 27.98 | 0.31 |
16 | 4096 | twoAwayXY, PMEPencils=8, lblUnload=yes | 12.93 | 0.58 |
16 | 4096 | twoAwayXY, PMEPencils=12, lblUnload=yes | 17.27 | 0.77 |
16 | 4096 | twoAwayXY, PMEPencils=16, lblUnload=yes | 16.02 | 0.72 |
16 | 4096 | twoAwayXY, PMEPencils=20, lblUnload=yes | 15.41 | 0.69 |
16 | 4096 | twoAwayXY, PMEPencils=12 | 16.21 | 0.73 |
16 | 4096 | twoAwayXY, PMEPencils=16 | 17.92 | 0.80 |
16 | 4096 | twoAwayXY, PMEPencils=20 | 17.99 | 0.81 |
16 | 4096 | twoAwayXY, PMEPencils=24 | 17.83 | 0.80 |
16 | 4096 | twoAwayXY, PMEPencils=36 | 16.97 | 0.76 |
8 | 4096 | twoAwayXY, PMEPencils=20 | 18.24 | 0.82 |
16 | 4096 | twoAwayXY, PMEPencils=20 | 17.99 | 0.81 |
32 | 4096 | twoAwayXY, PMEPencils=20 | 13.94 | 0.63 |
4 | 512 | 2.86 | 1.03 | |
8 | 512 | 2.84 | 1.02 | |
16 | 512 | 2.79 | 1.00 | |
32 | 512 | 2.29 | 0.82 | |
16 | 512 | ldbUnloadZero=yes | 2.79 | 1.00 |