Peak finding can gobble huge amounts of memory

DavidT3 / XGA

X-ray: Generate and Analyse is a module designed to make the analysis of XMM observations simple and efficient. It provides an interface with SAS for the creation of XMM data products, as well as a way to easily perform fits (scalable for multiple observations) and retrieve information about an object, all within a Python package.

BSD 3-Clause "New" or "Revised" License

29 stars 3 forks source link

Peak finding can gobble huge amounts of memory #626

Open DavidT3 opened 3 years ago

DavidT3 commented 3 years ago

Running the Sun et al. group sample on the XCS node, runaway memory usage occurs when peak finding for the group located at (18.214169 15.514934). You can sort of see why the peak finding isn't super effective on that image:

but because (I assume) the XCS node has 512Gb of RAM, it just runs away and would definitely fill it up if I let it. I think the behaviour in that case would be to raise an exception which XGA would catch and then it would carry on, but I'm a little cautious of letting it get that far.

I should identify the reason for this huge memory growth, and find a way to catch it before it happens

DavidT3 commented 3 years ago

When I let it grow it just hangs at 264GB, so I need a way of catching this

DavidT3 commented 3 years ago

Actually something odd is happening here - this behaviour happens at the same stage even though I removed the cluster that I thought was causing the problem.

DavidT3 commented 2 years ago

Its happened again!

This time with one of the ACT-DR5 clusters, really need to figure this out

DavidT3 commented 2 years ago

I really don't get why this happens - I think something about the huge memory capacity of the module must just let it take vast amounts.