miniwdl AWS plugin

Extends miniwdl to run workflows on AWS Batch and EFS

This miniwdl plugin enables it to submit AWS Batch jobs to execute WDL tasks. It uses EFS for work-in-progress file I/O, with optional S3 rails for workflow-level I/O.

The following assumes familiarity with local use of miniwdl run.

Use within Amazon SageMaker Studio

Start with the companion miniwdl-aws-studio recipe to install miniwdl for interactive use within Amazon SageMaker Studio, a web IDE with a terminal and filesystem browser. You can use the terminal to operate miniwdl run against AWS Batch, the filesystem browser to manage the inputs and outputs on EFS, and the Jupyter notebooks to further analyze the outputs.

That's the best way to try miniwdl-aws and get familiar with how it works with Batch and EFS. Read on for non-interactive deployment, which is a bit more complicated.

Unattended operations

For non-interactive use, a command-line wrapper miniwdl-aws-submit launches miniwdl in its own small Batch job to orchestrate the workflow. This workflow job then spawns task jobs as needed, without needing the submitting computer (e.g. your laptop) to remain connected for the duration. Separate Batch compute environments handle workflow & task jobs, using lightweight Fargate resources for workflow jobs. (See below for detailed infra specs.)

Submitting workflow jobs

First pip3 install miniwdl-aws locally to make the miniwdl-aws-submit program available. The following example launches a viral genome assembly that should run in 10-15 minutes:

miniwdl-aws-submit \
  https://github.com/broadinstitute/viral-pipelines/raw/v2.1.28.0/pipes/WDL/workflows/assemble_refbased.wdl \
  reads_unmapped_bams=https://github.com/broadinstitute/viral-pipelines/raw/v2.1.19.0/test/input/G5012.3.testreads.bam \
  reference_fasta=https://github.com/broadinstitute/viral-pipelines/raw/v2.1.19.0/test/input/ebov-makona.fasta \
  sample_name=G5012.3 \
  --workflow-queue miniwdl_workflow \
  --task-queue miniwdl_task \
  --fsap fsap-xxxx \
  --s3upload s3://MY_BUCKET/test \
  --follow

The command line resembles miniwdl run's with extra AWS-related arguments:

Unless --s3upload ends with /, one more subfolder is added to the uploaded URI prefix, equal to miniwdl's automatic timestamp-prefixed run name. If it does end in /, then the uploads go directly into/under that folder (and a repeat invocation would be expected to overwrite them).

Adding --wait makes the tool await the workflow job's success or failure, reproducing miniwdl's exit code. --follow does the same and also live-streams the workflow log. Without --wait or --follow, the tool displays the workflow job UUID and exits immediately.

Arguments not consumed by miniwdl-aws-submit are passed through to miniwdl run inside the workflow job; as are environment variables whose names begin with MINIWDL__, allowing override of any miniwdl configuration option (disable wih --no-env). See miniwdl_aws.cfg for various options preconfigured in the workflow job container.

Run directories on EFS

Miniwdl runs the workflow in a directory beneath /mnt/efs/miniwdl_run (override with --dir). The outputs also remain cached there for potential reuse in future runs.

Given the EFS-centric I/O model, you'll need a way to manage the filesystem contents remotely. Deploy an instance or container mounting your EFS, to access via SSH or web app (e.g. JupyterHub, Cloud Commander, VS Code server).

You can also automate cleanup of EFS run directories by setting miniwdl-aws-submit --s3upload and:

• --delete-after success to delete the run directory immediately after successful output upload
• --delete-after failure to delete the directory after failure
• --delete-after always to delete it in either case
• (or set environment variable MINIWDL__AWS__DELETE_AFTER_S3_UPLOAD)

Deleting a run directory after success prevents the outputs from being reused in future runs. Deleting it after failures can make debugging more difficult (although logs are retained, see below).

Security note on file system isolation

Going through AWS Batch & EFS, miniwdl can't enforce the strict file system isolation between WDL task containers that it does locally. All the AWS Batch containers have read/write access to the entire EFS file system (as viewed through the access point), not only their initial working directory.

This is usually benign, because WDL tasks should only read their declared inputs and write into their respective working/temporary directories. But poorly- or maliciously-written tasks could read & write files elsewhere on EFS, even changing their own input files or those of other tasks. This risks unintentional side-effects or worse security hazards from untrusted code.

To mitigate this, test workflows thoroughly using the local backend, which strictly isolates task containers' file systems. If WDL tasks insist on modifying their input files in place, then --copy-input-files can unblock them (at a cost in time, space, and IOPS). Lastly, avoid using untrusted WDL code or container images; but if they're necessary, then use a separate EFS access point and restrict the IAM and network configuration for the AWS Batch containers appropriately.

EFS performance considerations

To scale up to larger workloads, it's important to study AWS documentation on EFS bursting mode and performance monitoring. Like any network file system, EFS throughput limits can cause bottlenecks; and worse, burst credit exhaustion can effectively freeze a workflow.

Management tips:

• Monitor file system throughput limits and burst credits in the EFS area of the AWS Console.
• Store large datasets on EFS in advance (to the extent affordable), increasing its burst throughput and credits.
• Spread out separate workflow runs over time and/or across multiple EFS file systems.
• Temporarily provision higher throughput than available from bursting mode (24-hour minimum provisioning commitment).
• Configure miniwdl and/or AWS Batch to limit the number of concurrent jobs and/or the rate at which they turn over (see miniwdl_aws.cfg for relevant details).

Logs & troubleshooting

If the terminal log isn't available (through Studio or miniwdl-submit-awsbatch --follow) to trace a workflow failure, look for miniwdl's usual log files written in the run directory on EFS or copied to S3.

Each task job's log is also forwarded to CloudWatch Logs under the /aws/batch/job group and a log stream name reported in miniwdl's log. Using miniwdl_submit_awsbatch, the workflow job's log is also forwarded. CloudWatch Logs indexes the logs for structured search through the AWS Console & API.

Misconfigured infrastructure might prevent logs from being written to EFS or CloudWatch at all. In that case, use the AWS Batch console/API to find status messages for the workflow or task jobs.

Appendix 1: expected AWS infrastructure

Requirements:

1. VPC in desired region
2. EFS + Access Point providing uid=0 access
   • set --fsap or MINIWDL__AWS__FSAP to access point ID (fsap-xxxx)
3. Task execution:
   • Batch Compute Environment
     • spot instances (including AmazonEC2SpotFleetTaggingRole)
     • instance role
       • service-role/AmazonEC2ContainerServiceforEC2Role
       • AmazonElasticFileSystemClientReadWriteAccess
       • AmazonEC2ContainerRegistryReadOnly
       • s3:Get*, s3:List* on any desired S3 buckets
       • s3:Put* on S3 bucket(s) for --s3upload
   • Batch Job Queue connected to compute environment
     • set --task-queue or MINIWDL__AWS__TASK_QUEUE to queue name
4. Unattended workflow execution with miniwdl_submit_awsbatch:
   • Batch Compute Environment
     • Fargate resources
     • execution role
       • service-role/AmazonECSTaskExecutionRolePolicy
       • AWSBatchFullAccess
       • AmazonElasticFileSystemFullAccess
       • AmazonEC2ContainerRegistryPowerUser
       • s3:Get*, s3:List* on any desired S3 buckets
       • s3:Put* on S3 bucket(s) for --s3upload
   • Batch Job Queue connected to compute environment
     • tag the queue with WorkflowEngineRoleArn set to the execution role's ARN
     • set --workflow-queue or MINIWDL__AWS__WORKFLOW_QUEUE to queue name

Recommendations:

• Compute environments
  • Modify spot instance launch template to relocate docker operations onto NVMe or an autoscaling volume instead of the root EBS volume
  • Set task environment allocation strategy to SPOT_CAPACITY_OPTIMIZED and spot bid policies as needed
  • Set task environment spot bidPercentage as needed
  • Temporarily set task environment minvCpus to reduce scheduling delays during periods of interactive use
• EFS
  • Create in One Zone mode, and restrict compute environments to the same zone (trading off availability for cost)
  • Create in Max I/O mode
  • Configure EFS Access Point to use a nonzero user ID, with an owned filesystem root directory
• Use non-default VPC security group for EFS & compute environments
  • EFS must be accessible to all containers through TCP port 2049

Appendix 2: running tests

In an AWS-credentialed terminal session,

MINIWDL__AWS__FSAP=fsap-xxxx \
MINIWDL__AWS__WORKFLOW_QUEUE=WorkflowJobQueueName \
MINIWDL__AWS__TASK_QUEUE=TaskJobQueueName \
test/run_tests.sh

This builds the requisite Docker image from the current code revision and pushes it to an ECR repository (which must be prepared once with aws ecr create-repository --repository-name miniwdl-aws). To test an image from the GitHub public registry or some other version, set MINIWDL__AWS__WORKFLOW_IMAGE to the desired tag.