Got out-of-memory warnings while running

[HuoJnx]

Description:

I got some out-of-memory warnings while running, they are like this: I would like to know:

1. Is there any way to avoid them?
2. Will the warning affect a lot? Because I notice there are 25 (2+16+7) failures in total. But I have no concept about how many results will be affected.

Thank you very much for your help.

Input

A phyloseq object with 18777 OTUs and 40 samples.

Output

I write some little scripts to get the outputs. It seems that I still can get something from SPIEC-EASI.

Arguments:

pargs = list(rep.num=50, seed=1000, ncores=250) spiec_res=spiec.easi(phy_obj,method=“mb”,lambda.min.ratio=1e-3,nlambda=30,sel.criterion="bstars",pulsar.select=TRUE,pulsar.params=pargs)

Software version

SPIEC-EASI

#### R ## Computer #### CPU

Memory

[zdk123]

Using more than 50 cores actually doesn't do anything here, since we are parallelizing across the rep.num parameter. Try reducing the number of cores so that you can allocate more memory per thread. If that doesn't work maybe switch to batch.pulsar - that will write intermediate files to disk and computes the summary stat in a map-reduce framework, so it is much gentler on your RAM (at the expense of io)

[HuoJnx]

I understand, thank you very much for your help.

From: @.*** @.> on behalf of Zachary Kurtz @.> Date: Saturday, 27 August 2022 at 5:03 AM To: zdk123/SpiecEasi @.> Cc: HuoJnx @.>, Author @.***> Subject: Re: [zdk123/SpiecEasi] Got out-of-memory warnings while running (Issue #212)

Using more than 50 cores actually doesn't do anything here, since we are parallelizing across the rep.num parameter. Try reducing the number of cores so that you can allocate more memory per thread. If that doesn't work maybe switch to batch.pulsar - that will write intermediate files to disk and computes the summary stat in a map-reduce framework, so it is much gentler on your RAM (at the expense of io)

— Reply to this email directly, view it on GitHubhttps://github.com/zdk123/SpiecEasi/issues/212#issuecomment-1228945896, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AJXPY7UMCJ7XOBPXH7ZAN73V3EWJJANCNFSM5ZHEIIHQ. You are receiving this because you authored the thread.Message ID: @.***>