Skip to content
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

fhir_crack() | ncores > 1 takes unclear amount of RAM #114

Open
christiangierschner opened this issue Jul 21, 2022 · 2 comments
Open

fhir_crack() | ncores > 1 takes unclear amount of RAM #114

christiangierschner opened this issue Jul 21, 2022 · 2 comments

Comments

@christiangierschner
Copy link

when testing v2 with cracking 2600 bundles and various count of cores -> R took different amounts of RAM

e.g. ncores = 12 cracking went from 7,6 GB to 44 GB

i dont know if number of cores is related to amount of RAM taken

but so far performance is great :)

good job!

Christian

@palmjulia
Copy link
Collaborator

Hi Christian,

Thanks for submitting this issue. I would actually expect the RAM to increase when more cores are used because I think more or less the entire R session is copied into each fork, so more cores (=forks) -> more memory. The question is if we can find a way to reduce the memory load a bit. The first obvious attempt would be to keep the working space as small as possible when cracking a bundle list, i.e. make sure you don't have any other big objects lying around that you don't need at the moment. When I find the time I'll definetly look into this a bit more. Thanks!

Julia

@palmjulia
Copy link
Collaborator

palmjulia commented Aug 11, 2022

Update: The fork process used by R should do copy on write, so objects that are not touched during cracking should not be a problem.

I'm looking into possible solution from here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants