Accelerate development of bioinformatics workflows on AWS HealthOmics using call caching

This article explains how AWS HealthOmics call caching accelerates bioinformatics workflow development by reusing completed task outputs instead of rerunning entire workflows.

• Call caching saves completed task outputs to S3, enabling workflows to resume from failure points
• Two caching modes: "Cache on failure" for production runs, "Cache always" for rapid development iteration
• Supported across Nextflow, WDL, and Common Workflow Language workflow engines
• Tasks must define containers and produce outputs to be cached; non-deterministic tasks should be excluded
• Engine-specific controls available: Nextflow cache directive, CWL WorkReuse, WDL volatile attribute
• Reduces costs by eliminating redundant computation and accelerates development cycles significantly
• Provides debugging access to cached intermediate files and task execution artifacts
• Best practices include organizing caches by project, ensuring deterministic tasks, and managing S3 storage lifecycle

Call caching in AWS HealthOmics enables researchers to iterate faster on genomic workflows while reducing computational costs and time-to-insight for omics analyses.