Intel® Rack Scale Design Gaining Momentum

Intel® Rack Scale Design (Intel® RSD) is enjoying great traction across the ecosystem as more systems are fielded by major OEMs. Intel® RSD based products are currently available from many of the leading data center equipment suppliers, including:

Adding capabilities for storage disaggregation and composition, Quanta Cloud Technology, Supermicro, and Wiwynn have introduced high-performance, high-density pooled NVMe storage shared by multiple hosts over high-speed PCI-E interconnects. This enables customers to get the full performance advantage of Intel’s latest NVMe storage technology along with the flexibility provided by Intel RSD composability. In addition, Wiwynn is shipping the Wiwynn Cluster Manager* software, which provides resource composability, massive firmware and OS deployment, and real-time rack-level visual monitoring.

2018 will be an exciting year, as more vendors announce NVMe pooled storage over PCIe early in the year. Then in the spring, Intel will release specs and reference code to enable storage pooling using fast, scalable NVMe over Fabrics* interconnects. This provides an additional method for storage disaggregation with great performance and scalability extending across multiple racks. NVMe over Fabrics pooled storage products will follow in the second half of 2018.

Intel® RSD Customer Centers Coming on Line

Another big development is the opening of multiple Intel® RSD resource centers allowing customers to try RSD-based systems for themselves and to benchmark their critical workloads. The Dell EMC Solution Center* in Amsterdam enables customers to evaluate the DSS 9000 rack scale infrastructure via an on-site visit or remote login. Similarly, the Supermicro RSD Solution Center* in San Jose also supports in-person or remote evaluations. Wiwynn is participating in the Facebook Disaggregate Lab* with their SV7221G2 (Leopard)*, ST7200 (Lightning)*, and ST7110G2 (Knox)* RSD-based products (watch Facebook's Disaggregate Lab Virtual Tour video.) We expect to see many more labs open in 2018, giving customers a chance to learn about and put Intel® RSD through its paces.

Cloud Providers Getting Engaged with Intel® RSD

We’re seeing great interest from cloud software providers because Intel® RSD allows compute, storage, and accelerator sharing via resource pools, which can be “composed” into any desired configuration on the fly, adding a new level of agility to cloud-oriented data centers. Canonical*, a provider of Ubuntu*, has integrated Intel® RSD APIs into their Metal-as-a-Service (MaaS)* and Juju* toolsets, and also in their OpenStack* distribution. Red Hat* announced support for Redfish* APIs in OpenStack 12*, paving the way to a full Intel® RSD implementation in an upcoming release.

These integrations enable cloud software to control software-defined infrastructure via the Intel® RSD APIs, including the ability to compose systems on the fly to meet changing workload requirements. As a demonstration, Intel has enabled bare metal composition and provisioning from an OpenStack CLI using Intel RSD APIs. For more detail, read the Implementation Guide: "Provisioning Intel Rack Scale Design Bare Metal Resources in the OpenStack Environment."

At the CloudNativeCon/KubeCon* conference last December, SUSE and Intel demonstrated how Intel® RSD can also work with container environments like SUSE CaaS Platform* to deliver the benefits described above. The demo shows the SUSE CaaS (Containers-as-a-Service) Platform using Intel® RSD POD Manager (PODM) APIs to compose an optimized node incorporating fast non-volatile storage—i.e., NVMe drives connected over PCIe—for a container workload. The demo also shows how NVMe resources can be easily returned to resource pools, ready for a new workload. Read the Implementation Guide: "Dynamic Container Scaling with Intel Rack Scale Design," for a detailed description of the demo.

Telemetry Helps Reduce DC Energy Costs

New capabilities in the latest release of Intel RSD APIs and reference code include advanced out-of-band telemetry, which provides access to agentless monitoring and control features available on the latest Intel® Xeon® Scalable (Purley) platform. Customers are already using Intel RSD telemetry to implement advanced data center solutions like improved temperature controls to reduce energy consumption and to optimize workload placement. The latest reference code is available now on the Intel® RSD GitHub.Intel RSD APIs and reference code include advanced out-of-band telemetry, which provides access to agentless monitoring and control features available on the latest Intel Xeon Scalable (Purley) platform.

Coming in 2018

Looking forward to new developments this year, we will announce the next release of Intel® RSD in the spring. This will support storage pooling using NVMe over Fabrics interconnects, as mentioned above. The new APIs and code will also integrate comprehensive SNIA Swordfish* storage management standards as part of Intel® RSD within the RESTful Redfish framework. This will enable full discovery, provisioning, volume mapping, and monitoring of block, file, and object storage devices via the Intel® RSD APIs. An early release of the specs and reference code has already been sent to ecosystem partners.

In related news, AMI and Microsemi made a joint announcement of their collaboration on the development of an NVMe over Fabrics solution using AMI's MegaRAC FP-X Fabric Management* firmware, integrated with Microsemi's Switchtec* PCIe switches based on Intel® RSD reference architecture. As a side note, Supermicro and Intel demonstrated pooled storage using NVMe over Fabrics at the Gartner Datacenter* conference in December.

If you would like to learn more about Intel® RSD or get more involved in the Intel® RSD ecosystem, there are three great ways to engage with us. First, check out online information on the Intel® RSD website. Or you can contact your OEM or Intel sales representative now for an evaluation of Intel® RSD-based systems. If you are planning to attend Mobile World Congress or the OCP Summit, please visit Intel and our partners at these upcoming events to see Intel RSD for yourself.

Be sure to follow me on Twitter @UlgenFigen

Published on Categories Cloud ComputingTags , , ,
Figen Ulgen

About Figen Ulgen

Figen Ülgen, Ph.D., is General Manager of the Rack Scale Design Group within the Datacenter Group at Intel Corporation. Ülgen oversees solutions roadmap, solutions enablement as well as operations, support and ecosystem and partnership strategy for Rack Scale Design. Ülgen joined Intel in 2014, where she led as General Manager for HPC Software and Cloud developing and promoting HPC Platform Software technologies for the HPC ecosystem. She founded the award-winning OpenHPC Community under Linux Foundation with a group of ecosystem partners to democratize HPC. During her time with the HPC Software and Cloud, the team shipped Intel® HPC Orchestrator, a fully supported production quality software product in less than 18 months—concept to General Availability. Prior to that, she was a Senior Director with Microsoft for over a decade in various leadership roles in Data Science, Azure Public Cloud, and HPC teams. Her 20-year industry experience includes technology innovation, competitive strategy, software product planning and management, as well as driving change across large organizations. Before Microsoft, Ülgen worked at McKinsey & Co., Motorola and at Justsystem Corporation in Japan. With a passion for serving underprivileged youth and women in the technology field, Ülgen sits on the Saturday Academy Board of Directors and is regularly invited to speak on behalf of women in technology. Ülgen earned a Bachelor’s Degree in Computer Engineering, a Master’s Degree in Expert Systems under a Fulbright Scholarship and a Ph.D. in Machine Learning under the Japanese Government Ministry of Education Scholarship. Ülgen holds two U.S. patents.