Storage Developer Conference

Intel Fellow Amber Huffman has been at the center of Intel’s development for SSDs, with emphasis on SSD storage interfaces and next generation form factors. In this talk, she will discuss the rationale behind decisions that were made in advancing storage architecture leading to the emergence of solid state only Data Centers. Amber will also discuss the key factors influencing the future of the Data Center and the important role storage continues to play.

Join Dr. Andy Walker for a wonderfully illustrated tour through 90 years of physics and materials science, leading to modern solid state memory technologies via “The Golden Thread of Tunneling.”

NVMe KV is a proposal for a new command structure to access data on an NVMe controller that is being developed within the NVMe technical working group. This proposed command set provides a key and a value to store data on the Non Volatile media and provides a key to retrieve data stored on the media. In addition to the work on the NVMe specification, the SNIA is also working on a KeyValue API. This presentation will describe the standardization efforts going on in both the NVMe working group and SNIA.

Learning Objectives: 1) What is Key Value Storage; 2) Where are we at in the NVMe standardization process of NVMe KV; 3) Where are we at in the SNIA standardization process of a KV API?

The DMTF’s Redfish is an open industry standard specification and schema that specifies a RESTful interface and utilizes JSON and OData to help customers integrate solutions within their existing tool chains.

Designed to meet the expectations of end users for simple, modern and secure management of scalable platform hardware, the Redfish method and approach has seen accelerated adoption across the IT infrastructure management standards community.

This session provides an overview of the Redfish standard as well as information about what’s new in the Redfish ecosystem, including PLDM for RDE (Redfish Device Enablement) and other DMTF updates as well as consideration and adoption in the broader standards community. This session will then cover the general data model including the base storage models and infrastructure that are used by the SNIA Swordfish extensions.

Learning Objectives: 1) Basic understanding of the Redfish data model hierarchy; 2) Knowledge of the problem areas to which the Redfish methodology is being applied; 3) Understand how RDE brings Redfish “inside the box” for network & other devices.

As data-intensive workloads transition to low-latency NVMe flash-based storage to meet increasing user demand, the Fibre Channel industry is combining the lossless, highly deterministic nature of Fibre Channel with NVMe. FC-NVMe targets the performance, application response time, and scalability needed for next-generation data centers while leveraging existing Fibre Channel infrastructures. This presentation will provide an overview of why Fibre Channel’s inherent multi-queue capability, parallelism, deep queues, and battle-hardened reliability make it an ideal transport for NVMe across the fabric.

Learning Objectives: 1) A reminder of how Fibre Channel works; 2) A reminder of how NVMe over Fabrics work; 3) A high-level overview of Fibre Channel and NVMe, especially how they work together.

Nantero NRAM™ is a new class of memory with the potential to add non-volatility to existing RAM applications. It can be arranged in a crosspoint structure for large memories or a 1T-nR arrangement for smaller faster arrays, in standalone devices or as embedded RAM. NRAM uses carbon nanotubes in a dielectric-free structure to achieve unlimited write endurance. While there are obvious advantages to this class of device, including replacing DRAM in storage devices, there are a number of less obvious changes to how designers approach the data storage hierarchy. Decoupling cache size from battery backup power lets designers rethink performance profiles. Exploitation of various interfaces into the system are examined, from SATA to PCIe to the many options in the DRAM bus. This presentation also explores the growing application space for artificial intelligence, deep learning, and in-memory computing and considers the impact of a high performance non-volatile memory in those use cases.

Learning Objectives: 1) Awareness of performance characteristics of a DRAM-class non-volatile memory in crosspoint and 1T-nR configurations; 2) Non-obvious use cases for NRAM; 3) Application of non-volatile memory in AI / deep learning engines.

Leveraging the NVMe standard to present FPGA accelerators as NVMe namespaces allows user space code to access accelerators as simple block devices via standard in-box drivers. NVMe-over-Fabrics (NVMe-oF) is a network protocol for NVMe that is used to communicate between a host machine and NVMe devices over high-performance Ethernet networks. Leveraging NVMe-oF allows an accelerator, that presents as a standard NVMe namespace, to be shared across existing transports such as RDMA, TCP/IP or Fibre Channel. With NVMe-oF, client machines can borrow accelerators over the fabrics connection which can then be accessed as if they were local to the system, allowing the same user code to run on direct attached accelerators and over-Fabrics accelerators. The ease and granularity available when setting up the over-fabrics connections allows servers to share accelerators on demand, allowing for the disaggregation of acceleration compute resources. Using NVMe Controller Memory Buffer (CMB), the fabrics connection is further improved by allowing devices to send and receive data directory from the accelerator without accessing host memory. We will discuss the advantages of accelerator disaggregation and its impact on resource usage.

Learning Objectives: 1) Advantages and performance of NVMe for storage workload acceleration; 2) Performance of NVMe over Fabrics for storage workloads and acceleration; 3) Advantages of using Controller Memory Buffer coupled with NVMe over Fabrics.

Cloud-native frameworks are becoming a de-facto choice for deploying scalable micro-services. Flash based storage manageability in the container frameworks need to evolve to deliver scalable storage services to container workloads. OpenSDS (Linux Foundation project) aims to provide a unified software-defined storage control plane to simplify flash storage manageability to service on-prem and cloud micro-services. In this session, we present NVMe and NVMe over Fabrics (NVMe-oF) manageability architecture in OpenSDS using Swordfish APIs, a case study with Kubernetes and community roadmap to deliver unified flash manageability vision.

Learning Objectives: 1) Flash manageability using Swordfish APIs; 2) OpenSDS Linux Foundation project overview (unified storage control plane); 3) Container flash driver integration; 4) Kubernetes.

The SNIA’s Scalable Storage Management Technical Work Group (SSM TWG) has developed an open industry standard specification that provides a unified approach for the management of storage systems and data services. Swordfish™ is an extension of the DMTF Redfish specification developed by the Storage Networking Industry Association (SNIA) to provide a unified approach for the management of storage equipment and services in converged, hyper-converged, hyperscale and cloud infrastructure environments, making it easier for IT administrators and DevOps to integrate scalable solutions into their data centers.

This session will present an overview of the SNIA Swordfish specification, and will show how Swordfish takes and extends the Redfish specification to deliver the Swordfish storage model. It will also cover the drivers for the SNIA Swordfish approach, as well as providing a comprehensive overview of the functionality included in the Swordfish specification. It will also present an overview of common storage management use cases easily addressed by Swordfish.

Learning Objectives: 1) Understand why and how SNIA partnered with DMTF to create Swordfish; 2) Learn about the growing importance of Swordfish to the storage industry; 3) Get pointers to online Swordfish resources, communities and tools; 4) Discover storage management use cases for data center equipment and services.

NVMe over Fabrics is a powerful standard that provides fast access to non-volatile memory devices across fabric interconnects. An emerging NVMe-oF transport is good old TCP/IP. Its benefits are obvious, as it is fast, scalable, well-understood, and extremely simple to deploy. TCP/IP is, after all, the most widely used network protocol of them all, and well known and heavily-implemented in every data center. The major question is how to achieve high performance and low latency with TCP/IP. As most designers know, TCP/IP has pitfalls, such as timeout-based retranmissions and incast. However, performance results are very promising. The NVMe Technical Working Group is building both the standard and its open-source implementation in concert. Its prominent features include data integrity and transport layer security. In practice, NVMe/TCP is an excellent transport for networked flash and complements NVMe/RDMA and NVMe/FC nicely.