Crunch - container orchestration

Arvados has a robust container orchestration system called 'Crunch', which executes CWL workflows while maintaining provenance and reproducibility.

Design Goals

Notable design goals and features include:

  • Scale horizontally to support very large workloads
  • Integrate with Keep and git repositories to maintain provenance
  • Use off-the-shelf software tools in distributed computations
  • Efficient over a wide range of problem sizes
  • Maximum flexibility of programming language choice
  • Maximum flexibility of execution environment
  • Low entry barrier for bioinformaticians
  • Isolate workloads by running all jobs inside Docker containers

Benefits of Crunch

Although some of the workflow and provenance features in Arvados could theoretically be implemented using Hadoop MapReduce, there are distinct benefits to Crunch:

  • Provenance and Reproducibility - Like Keep, the Arvados distributed file system, Crunch is designed to automatically track the origin of result data. It can also efficiently reproduce complex workflows and comparing workflows to one another.
  • Performance - Most genomics problems are embarrassingly parallel and can benefit from horizontal scaling. In the cloud, Crunch can deliver cost-effective performance for genomics related analyses by automatically adjusting the available compute resources to the workload.
  • Standardization - Common Workflow Language is the workflow description standard in bioinformatics. It is the native workflow language in Crunch.