Parabricks 4.3.1-1 htvc-low-memory usage

[edwin_mdu]

I tried to solve cudaSafeCall() failed - out of memory in T4 GPUs (g4 instances) on AWS healthomics workflows by enabling htvc-low-memory. But I get an error as “check JSON input; unknown input/output: htvc-low-memory. WDL engine failed” May I know the mode “htvc-low-memory” usage is different now in 4.3.1-1.

The following is my inputs.json

{
“inputFASTQ_1”: “s3://mybucket/myfastq1.fastq.gz”,
“inputFASTQ_2”: “s3://mybucket/myfastq2.fastq.gz”,
“inputRefTarball”: “s3://mybucket/my_assembly38.fasta.tar”,
“htvc-low-memory”: true
}

Thanks

[edwin_mdu]

Please ignore the above and you may close the ticket…AWS Healthomics wdl workflow is based on parabricks docker 4.1.1-1 . So the workflow parameter don’t have htvc-low-memory tag option…