Organization: Lawrence Berkeley National Laboratory
Year: 2016

Shifter is an open-source tool to improve the flexibility and usability of HPC systems for data-intensive workloads by leveraging emerging container-based computing tools such as Docker to enable supercomputer users to run a wider range of software more easily and securely. Shifter is focused on the needs and requirements of the HPC workload, which means it delivers the functionality users are looking for while still meeting the overall performance requirements and constraints that a large-scale HPC platform and shared resource providers impose. But Shifter is not just for the largest HPC centers—it is also valuable for departmental and university clusters where small to moderate size clusters are used by a variety of researchers. Shifter can enable even these smaller systems to be used more flexibly. As container-based computing gains popularity, many research communities are starting to define and share images that contain integrated and tested bundles of software optimized for specific applications. Images exist for assembling genomic sequences or analyzing telescope images. Researchers will want to use these containers on their shared resources, and Shifter provides an avenue to support this.