-
Notifications
You must be signed in to change notification settings - Fork 29
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
Stop memory consumption increasing linearly with time #1456
Comments
If we instrument the eicrecon-dis job here with prmon, does it show the same behavior? I.e. can we set something up that tracks this and flags degradation? |
Entirely anecdotal, but I don't notice this on my own system which has a few differences from the containers we're running here, notably jana2@master and [email protected]. |
I'm not able to reproduce with my local setup either (JANA2 2.2.1-rc1, podio 0.99).
@rahmans1 Can you run with |
That was for single particle ^^ I ran a long job over pythia8CCDIS_10x100_minQ2=1000_beamEffects_xAngle=-0.025_hiDiv_1.hepmc3.tree.root
Can't see a leak, still. |
Re-running with npsim in the 24.05.0-stable container, I'm able to reproduce:
The upper frame looks like:
|
Reproducer:
|
libIRT is leaky: eic/irt#36 IIRC, some |
Is your feature request related to a problem? Please describe.
For a wide range of datasets in simulation production eicrecon shows linear increase in memory consumption over time. The plot below demonstrates an example of this issues by comparing the memory consumption over time in eicrecon vs npsim for one of these datasets.
pythia8CCDIS.pdf
Describe the solution you'd like
We would prefer the rss to be under 2 GB in eicrecon for the duration of the run. This will allow us to achieve better throughput in Open Science Grid resources. The first step would be to identify why the memory consumption increases linearly without any obvious signs of memory leak.
Describe alternatives you've considered
Additional context
The text was updated successfully, but these errors were encountered: