Why is Cromwell launching "n1-ultramem-40" GCE instance when I ask for 'memory: "16 GB" cpu: 36'?

tomdyartomdyar Member

I am running the Broad's viral-ngs (http://viral-ngs.readthedocs.io/en/latest/index.html) "demux_plus.wdl" pipeline, on Cromwell version 31 via docker-compose with mysql database, with this runtime section for the first step:

runtime {
docker: "quay.io/broadinstitute/viral-ngs"
memory: "16 GB"
cpu: 36
dx_instance_type: "mem1_ssd2_x36"
preemptible: 0 # this is the very first operation before scatter, so let's get it done quickly & reliably

Here are the overall defaults:

default-runtime-attributes {
cpu: 2
memory: "4 GB"
failOnStderr: false
continueOnReturnCode: 0
bootDiskSizeGb: 50
# Allowed to be a String, or a list of Strings (NB, was "LOCAL" instead of "HDD")
disks: "local-disk 500 HDD, /mnt/tmp 500 HDD"
noAddress: false
preemptible: 1
zones: [ "us-central1-a", "us-central1-b", "us-central1-c", "us-east1-b", "us-east1-c", "us-east1-d" ]

For some reason, the instance type requested is huge: "n1-ultramem-40 -- Memory-optimized machine type with 40 vCPUs and 961GB of system memory". Could this be a bug? Can I set some debugging output to get some info on why this instance type was chosen? Do I need to specify a different "actor-factory" class other than "cromwell.backend.impl.jes.JesBackendLifecycleActorFactory"?



Best Answer


Sign In or Register to comment.