Gary Grider will kick off the talk with a brief background of the challenges currently faced in mass storage systems, and why it is so difficult for modern S3 based workloads to utilize them. Ben McClelland, who architected the system and did the bulk of the development, will introduce the Versity Gateway, an open-source high-performance object-to-file translation tool. He will explain that the Versity Gateway was written from scratch in Go, highlighting the benefits of this choice.

This talk presents the design and implementation of libvfn, a new open-source library for interacting with PCIe-based NVMe devices from user-space using VFIO. The core of the library is excessively low-level and aims to allow NVMe controller verification and testing teams to interact with devices at the register and queue level. While the library ships with a production ready NVMe driver with a high-level API, the library is designed to expose enough low-level VFIO functionality with which custom drivers can be implemented for any PCIe device as required by the application.

As large deployments become more common, our industry needs a standardized, multi-vendor solution to boot computer systems from OS images stored on NVMe® devices across a network. The newly published NVM Express® Boot Specification and ecosystem partnership with UEFI and ACPI enables this by leveraging the NVMe over Fabrics (NVMe-oF™) standard. This talk is a dive into the details of the new specification and the design of an open-source prototype for booting over NVMe/TCP transport using a UEFI implementation.

Deoxyribonucleic Acid (DNA) as a storage medium with high density and long-term preservation properties can satisfy the requirement of archival storage for rapidly increased digital volume. The read and write processes of DNA storage are error-prone. Images widely used in social media have the properties of fault tolerance which are well fitted to the DNA storage.

A new error correction code for DNA data storage is presented. The code takes advantage of the information about the noise in DNA data channel to clean up errors and erasure, that's why it is called DNA Noise Aware Errors Erasures Cleaner (DNAe2c). By modeling errors and erasures source with different state-of-the-art distributions and real data, we see an improvement of 10x over Reed Solomon codes requiring less than 20% overhead, making DNAe2c a promising candidate to accelerate DNA data storage adoption.

Synthetic DNA-based data storage has been on the rise as a candidate for Data Storage due to its longer shelf life and higher data density. This technology is expected to tackle the ever-increasing demand for cold storage and reduce energy consumption of data-centers to preserve information over long periods of time. In 2021, Lenovo and IPT joined the race towards DNA synthesis for data storage application. In regards to the traditional chemical method, the established “base-by-base” synthesis promises the highest information density on short length DNA strands.

There are several well-known advantages of using synthetic DNA for cold-data storage, such as higher density, reduced energy consumption, and durability compared with the standard storage mediums used for the same purpose. The enablement of this technology in the market involves the development of cost-effective DNA synthesizers that can write the data at an appropriate throughput speed and a CODEC able to handle data from different synthesis and sequencing technologies.

Security for NVMe over Fabrics is highly sought by customers and evolving at a fast pace. Experience with initial implementations uncovered a few shortcomings in the original design of certain NVMe/TCP security features that have resulted in recent security protocol changes. Furthermore, a lack of understanding on how configured single-system NVMe over Fabrics security policies interact led to the standardization in NVM Express of an interoperability framework. This session will explain these recent developments in NVMe-oF security and their importance to achieve a secure SAN.

Almost 10 years ago, the SPDK userspace polled mode nvme driver showed performance and efficiency far surpassing what was capable with the Linux kernel. But in recent years, Linux has responded with io_uring and asynchronous NVMe passthrough interfaces. The xNVMe project has also helped storage projects and applications adapt to the ever-growing list of Linux storage interfaces. This talk will compare the strengths of the SPDK and Linux NVMe drivers, explain how xNVMe has enabled io_uring NVMe passthrough in SPDK, and share some early performance results.

Firmware update, security management and initial system provisioning need to be performed at scale, which is a challenge for many environments. Standards-based Management is essential for interoperability between the components required for large-scale deployments. Standards-based out-of-band mechanisms such as MCTP and NVMe-MI provide the basic functionality but need Swordfish to provide the client interface and scalability. Implementations have been developed as open source and upstreamed to integrate between these two layers.