Datashim is enabling and accelerating data access for Kubernetes/Openshift workloads in a transparent and declarative way. Opensourced since September of 2019 and is growing to support use-cases related to data access in AI projects. It brings benefits across different entities:
- Data scientists/engineers: Focus on workload/experiments development and not on configuring/tuning data access
- Storage Providers: Increase adoption since the framework is extensible without hindering the User Experience
- Data-oriented Frameworks: Can build capabilities (caching, scheduling) on top of DLF using a declarative way to access/manage data sources
Datashim is a incubation-stage project of the LF AI & Data Foundation.
Contributed by: IBM in January 2021