Community discussion of Storage for AI Workload Definitions, Characterization, and Benchmarking needs and wants

Critical to hardware and software development is an understanding of the problems being addressed. The unrelenting evolution of AI impacts the supporting infrastructure just as much as the AI accelerators that have been the focus of development. Multiple efforts are underway to characterize the load and requirement for storage devices and systems in AI environments, each with different specific goals.

This Birds of a Feather will bring members of industry working groups together to discuss the current state of storage for AI workload characterization, benchmarking, tracing, testing, and validation then open the floor to attendees to identify gaps and opportunities so that these industry efforts can evolve at the same pace as AI.