Paper Title: EDSFF for NVMe Computational Storage Processors

Paper Abstract: The EDSFF specification addresses the challenges of deploying NVM and accelerators in next-generation data-centers. The form-factor is incredibly well suited for the deployment of NVMe Computational Storage Processors which can be used to perform storage and analytic acceleration close to NVMe SSDs. By deploying these NVMe CSPs on E1.S applications can run faster and more efficiently than ever before. They reduce energy costs via CPU offload and reduced data-movement. An example of a E1.S based NVMe Computational Storage Processor will be presented and its benefits to end users will be discussed.

Paper Author: Stephen Bates, VP and Chief Architect of Emerging Storage Systems, Huawei

Author Bio: Stephen Bates is CTO at Eideticom, a developer of leading edge storage, compute, and applications for programmable platforms in the cloud or at the network edge. He focuses on applying emerging technologies such as NVMe, RDMA, new non-volatile memories, and advanced programmable logic to create complex storage and communications systems. He has combined several such technologies to implement computational storage that offers performance well above today’s production systems. He is also an active contributor to the Linux kernel. Before joining Eideticom, he worked in the CTO office at PMC-Sierra and was a professor of computer engineering at the University of Alberta. He holds a PhD in signal processing from the University of Edinburgh, Scotland. He has given presentations at Storage Developer Conference and at past Flash Memory Summits.