-
Notifications
You must be signed in to change notification settings - Fork 0
Add parallelism and temperature to water entropy method call. #129
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add parallelism and temperature to water entropy method call. #129
Conversation
This will trigger water entropy to run multithreaded. We should later percolate parallel worker threads into water entropy from CE to save overhead in setting up two lots of schedulers.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This looks great! I can see that waterEntropy
is now running in parallel, which should lead to a significant speedup in CodeEntropy
. Thanks very much for implementing this!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Included the extra dictionary produced from v1.2.0 of waterEntropy and it works on the arginine example trajectory. Parallel is switched on by default, which is fine, but can be included as an argument in the config file if requested in the future.
We can add it as a configurable parameter. I was initially going to leave it until it is clear whether or not we are implementing parallelism in CE. Since at this point I would initiate the worker topology in CE and pass the context into WE to override parallel setup. It's not clear yet if the slowness is coming from the conformational entropy added lots of pointless empty lists, and once fixed makes it fast enough to not bother. What does every one think? @jkalayan @harryswift01 @skfegan Expose parallel/serial to config? I'm not inclined to either way so if there is an opinion then please put it forward and we will implement. Also I think I need to fix at least one test before I merge this. |
Fixed the tests now |
Summary
This PR is to include the parallel and temperature parameters to the waterentropy function call such that the default temperature set in waterentropy is overridden by that set in CE config file and also to cause water entropy to run in parallel.
Changes
Update function call to waterentropy with temperature and parallel params:
Impact
This will make sure that temperature in waterentropy is set to the CE value and will trigger water entropy to run in parallel. We still need to parallelise CE but when we do the context should be passed also into WE so that we avoid the overhead of dual dask setups.
closes #125