rpetit3 / dragonflye

:dragon: :fly: Assemble bacterial isolate genomes from Nanopore reads
GNU General Public License v3.0
117 stars 10 forks source link

Memory calculation #16

Closed lskatz closed 6 months ago

lskatz commented 1 year ago

I'm wondering if there is a way to estimate the memory requirements up front so that it doesn't produce an error. Is that even possible? Thinking about pilon but it's possible something else on here takes more memory? If it's insufficient memory, then I'd suggest dragonfly produces an up front error.

rpetit3 commented 1 year ago

I think with a lot of runs we could potentially estimate the amount of memory needed.

Curious, what size genome and estimated short read coverage was giving you issues?

lskatz commented 1 year ago

Default 16G memory and then the assembly size is around 4.8M. I'm not sure what the original coverage is. Trying to rope in my colleague to get all the details :)

rpetit3 commented 6 months ago

Hi @lskatz

I'm thinking I will let workflow managers and pipeline developers handle this. If you've come up with a solution for this or would like to revisit, please feel free to reopen.

Cheers, Robert