Keating:FutureCode: Difference between revisions
From OpenWetWare
Jump to navigationJump to search
No edit summary |
No edit summary |
||
| Line 5: | Line 5: | ||
* Get to the bottom of the apparent MPI slowdown between parent and children processes - a better timing study. | * Get to the bottom of the apparent MPI slowdown between parent and children processes - a better timing study. | ||
* Implement exclusion list capability (in PDB.pm probably) and allow the possibility of moving self calculations outside of CHARMM. Start by moving vdW such that arbitrary modifications can be easily applied. | |||
Revision as of 06:17, 29 October 2005
| Information concerning the Keating Lab | |
| Research | |
|
Resources: | |
- Allow the parent process in specdesign to be able to run an arbitrary command (specified in the control spec file) that re-evaluates the energy of the solution. Some ideas:
- popen() is a useful function in C.
- the specified command (most conveniently a script) must expect exactly one input string on the command line (the solution/backbone string) and output only one number is to stdout - the new energy.
- Get to the bottom of the apparent MPI slowdown between parent and children processes - a better timing study.
- Implement exclusion list capability (in PDB.pm probably) and allow the possibility of moving self calculations outside of CHARMM. Start by moving vdW such that arbitrary modifications can be easily applied.
