Design large warehousing solution

[krobasky]

Currently, redis queue data are persisted in mongodb container while large, expression, phenotype, and results datasets are persisted on disk.

close this ticket when:

• [ ] a release process has been tested and documented for persisting the mongodb job data across deployments
• [ ] large datasets are stored in a scalable, versionable fashion that adheres to FAIR principles (federated data features of iRODS may be helpful)
• [ ] any API design changes required for supporting the FAIR principles (e.g., data object versioning) have been addressed in the dashboard