Computational Storage

SNIA and the Compute, Memory, and Storage Initiative technical work and marketing iniitiatives are part of an exciting direction of adding computation to the normal storage features of devices and systems to realize a significant performance and infrastructure scaling advantage.

The industry is seeing an increase in customer requirements to move compute closer to traditional storage devices and systems.  Computational Storage solutions typically target applications where the demand to process ever-growing storage workloads is outpacing traditional compute server architectures. These applications include AI, big data, content delivery, database, machine learning and many others that are used industry-wide.  

The SNIA Computational Storage Special Interest Group (SIG) evangelizes SNIA's technical work in computational storage, performed by the SNIA Computational Storage TWG.  The TWG currently has out for public review the Computational Storage Architecture and Programming Model v0.3 rev 1, which defines recommended behavior for hardware and software that supports computational storage.  The recent SNIA Podcast, "What Happens When Compute Meets Storage?" provides highlights of their work.  Download the slides here.