Skip to content

Conversation

@afgane
Copy link
Contributor

@afgane afgane commented Nov 26, 2024

This includes resource requirement for the Galaxy-ME tutorial: https://training.galaxyproject.org/training-material/topics/imaging/tutorials/multiplex-tissue-imaging-TMA/tutorial.html

Values come from the cancer.usegalaxy.org instance.

tools.yml Outdated
_JAVA_OPTIONS: -Xmx{int(mem)}G -Xms1G
toolshed.g2.bx.psu.edu/repos/imgteam/bfconvert/ip_convertimage/.*:
cores: 12
mem: 128
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This seems to be a lot, is this really needed? Is this dependent on the input size?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I guess you haven't seen mesmer yet...

As I mentioned in the initial comment, the values come from the cancer.usegalaxy.org instance where the tools run correctly. So I figured we start there and bring it down if possible once we have some runtime data?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We do have this tool installed:

(venv) galaxy@sn06:~$ gxadmin tsvquery tool-metrics %ip_convertimage% memory.max_usage_in_bytes --like | awk '{print $1 / 1024 / 1024 / 1024}' | gxadmin filter histogram

(   0.114,   22.214) n=1433  **************************************************
[  22.214,   44.314) n=6     
[  44.314,   66.415) n=20    
[  66.415,   88.515) n=1     
[  88.515,  110.615) n=7     
[ 110.615,  132.716) n=1     
[ 132.716,  154.816) n=0     
[ 154.816,  176.916) n=0     
[ 176.916,  199.017) n=0     
[ 199.017,  221.117) n=1     
[ 221.117,  243.218) n=0     
[ 243.218,  265.318) n=0     
[ 265.318,  287.418) n=0     
[ 287.418,  309.519) n=0     
[ 309.519,  331.619) n=0     
[ 331.619,  353.719) n=0     
[ 353.719,  375.820) n=0     
[ 375.820,  397.920) n=0     
[ 397.920,  420.021) n=0     
[ 420.021,  442.121) n=0     
[ 442.121,  464.221) n=0     
[ 464.221,  486.322) n=0     
[ 486.322,  508.422) n=0     
[ 508.422,  530.522) n=1     

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OK, nice. So do you think we should go with 64GB or bring it all the way down to 24?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

At least this is what data suggesting, maybe you have large inputs and it takes more mem? or it scales with cores? In that case we need to have a nice rule.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have no usage data atm so I'll just drop it to 30GB for the time being.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Forgot I didn't actually update the config following my reply... Updated now.

@nuwang nuwang requested a review from mira-miracoli February 12, 2025 16:57
@nuwang
Copy link
Member

nuwang commented Feb 12, 2025

@bgruening Is this looking ok to merge?

@bgruening bgruening merged commit 1d3c48b into main Feb 12, 2025
1 check passed
@bgruening bgruening deleted the absolute-elk branch February 12, 2025 19:39
@bgruening
Copy link
Member

Thanks!

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

Successfully merging this pull request may close these issues.

4 participants