Do y'all have a timeframe on when you'll be releasing the Cromwell AWS backend?
We're targeting by end of August.
We are using AWS Batch as well for our pipelines. I'm interested in using Cromwell with AWS Batch (both WDL and CWL). Are there any best practices for using Cromwell with AWS Batch? Or specifics on containerizing applications? How does Cromwell handle transferring input data (folders and not just files) from S3?
Let's take bcl2fastq...My workflow would run bcl2fastq. The input would be the s3 path to the run folder, and the output would be a series of FASTQ files along with the demux qc metrics json file. If I use an s3 path to a folder, how will Cromwell know to download the whole run folder?
Additionally, how do I specify the resource requirements (cores, RAM, etc) required for a particular task? (maybe I didn't dig deep enough into the docs, or missed it...)