AWS HealthOmics introduces resource fallback order for Nextflow workflows
AWS HealthOmics now supports resource fallback for Nextflow workflows, allowing automatic switching between preferred accelerators or CPU instances to reduce downtime and manual resubmissions.
AWS HealthOmics has added a resource fallback directive for Nextflow workflow tasks, enabling users to specify an ordered list of preferred accelerator types. This feature allows researchers to prioritize accelerators while automatically falling back to alternatives if the preferred option is unavailable, reducing manual intervention. The directive supports configurable timeouts for each accelerator profile, giving users control over how long HealthOmics searches for a specific accelerator before moving to the next in the list.
The fallback order can include a CPU instance as a final option, ensuring tasks have a higher chance of reserving an instance even if accelerators are constrained. This feature is designed to minimize delays caused by accelerator shortages, particularly in production workflows where uninterrupted execution is critical. Users can now define fallback behavior without needing to resubmit runs manually.
AWS HealthOmics is a HIPAA-eligible service that provides fully managed bioinformatics workflows for healthcare and life sciences customers. The new resource fallback capability enhances the service’s ability to support large-scale scientific workflows by improving reliability and reducing operational overhead. Researchers can now focus more on analysis and less on troubleshooting accelerator availability.
The resource fallback feature for Nextflow workflows is available in all AWS HealthOmics Regions, including US East (N. Virginia, Ohio), US West (Oregon), Europe (Frankfurt, Ireland, London), Israel (Tel Aviv), and Asia Pacific (Seoul, Singapore, Tokyo). Users can configure and deploy fallback orders through the advanced resource configuration and Nextflow-specific documentation.