Introducing the Need for NFS-SSD the Ethernet Direct Attached SSD that Natively Speaks Network File System


As the one of the inventors of NVMe at Fusion-io, David has long been a thought leader in the SSD space. Large steps forward were made in data processing when NVMe was embedded in the server, bringing large quantities of high-performance data into direct contact with processing. However, data driven workload requirements have since changed, dramatically: 1) Larger quantities of data are being created, analyzed, and processed; 2) Demands for performance are not just growing, but, accelerating; and, 3) Most importantly, data needs to be more usable. Workflows, AI and ML engines, and applications need data generated at the edge, in datacenters, and in the cloud to be aggregated and shared over a network or a fabric. This has become critical in today’s decentralized workflows that are rapidly on the rise with the massive adoption of cloud computing, the increase in edge locations, and the large number of data driven employees working in remote locations. David will discuss why NVM over Fabric was destined to have minimal adoption and will put forth the assertion that we need to embed NFS in the device.

David Flynn
Related Sessions