Issue with data or with cluster.split command?

Hi all,

I am running data from 58 samples through the mothur pipeline. Everything seems to be going fine until the cluster.split step. Every time I’ve left it to run, I come back and my computer has shut off (despite telling it not to do this while plugged in, and I can see that there are many .list files, but the one that should be used in the make.shared step seems to be empty? Is this an issue with computing power/RAM so it’s shutting off before completion?

Everything seems to be running fine to this point (number of seqs are comparable to a colleague doing a similarly sized group. I’m stumped. Please help!

Hi bmd2791
Probably the mothur people will know better to answer your question but it sounds to me that there may be sth worng with your machine… Shutting down during high cpu usage could be because it overheats and it turns off to prevent any damage. Have you tried to run any programs that monitor CPU usage and components temperature while running cluster.split. That way you would be able to see whats happening inside your pc while it shuts down. Also i assume you are using your laptop, are you sure it is charging properly and it does not run out of juice while computing, cause if that happens then shutting down is also normal. Also, what the mothur.log file say when it crashed, any error messages? I would definitely try to run the command and keep an eye on it to see exactly whats going on when it shuts down
Good luck

How much RAM does your computer have? Can you possibly run it on a linux cluster (eg. AWS)? I think windows might limit the amount of RAM each process can utilize and so you’re likely running into that. Alternatively, you might also try reducing the number of processors you are using.