Spin

Spin services are built with Docker containers that can easily access NERSC systems and storage.

The system underlying Spin, Rancher, is a management and orchestration framework for Kubernetes clusters. The current Rancher system offers many of the same features that Kubernetes users will find familiar, but with an intuitive web-based user interface. Provided alongside Rancher is a container image registry based on the popular Harbor system.

nersc spin transp 640x460 mark only

Spin is a container-based service for NERSC users, in cloud terms, a containers-as-a-service, or CaaS platform. 

Researchers use Spin to deploy their own science gateways, workflow managers, databases, API endpoints, and other network services to support scientific projects.

Spin services are built with Docker containers that can easily access NERSC systems and storage.

The system underlying Spin, Rancher, is a management and orchestration framework for Kubernetes clusters. The current Rancher system offers many of the same features that Kubernetes users will find familiar, but with an intuitive web-based user interface. Provided alongside Rancher is a container image registry based on the popular Harbor system.

How scientists use Spin

The flexibility of Docker makes Spin useful for many projects. A few examples include the following:

  • The Dark Energy Spectroscopic Instrument (DESI) project operates databases and tools for data movement that support continuous workflows.

  • ESS-DIVE is a data repository that facilitates the sharing and discovery of earth sciences data sets using the Metacat research data catalog system on top of NERSC storage.

  • The LZ Dark Matter Experiment has built custom workflow tools and science gateways that help collaborators organize the experiment and analyze and visualize notable detector events.

  • Phytozome, a project of the Joint Genome Institute, offers a comprehensive library of plant genomics data.

  • ScienceSearch provides deep search into research data sets by generating robust metadata from ML techniques that incorporate user tagging and training.

NERSC also uses Spin as a front-end for some of its most popular services, such as JupyterHub.