Cavium Technologies powers end-to-end NVMe over Fabrics offerings; expands partnerships to push OCP NIC 3.0



Cavium announced on Monday that it is collaborating with Microsemi and Marvell to demonstrate a reference architecture for accelerating and scaling out NVMe over Fabrics (NVMe-oF). The architecture will be demonstrated live on ARM64 and x86 architectures at the OCP Summit 2018 in San Jose, California.

Integrating a broad range of proven Cavium technologies (ThunderX2 processors, FastLinQ 100GbE NICs, XPliant Programmable Switches) with Microsemi’s Switchtec PCIe switches and Flashtec NVRAM cards and Marvell’s NVMe SSD controllers enables multiple differentiated and validated end-to-end NVMe solutions, including NVMe-based all-flash arrays (AFAs), just a bunch of flash (JBOF) and fabric-attached bunch of flash (FBOF).

In addition to a hardware solution, the collaboration enables software integrations of Microsemi’s open source PCIe peer-to-peer (P2P) technology and Cavium FastLinQ concurrent RoCE and iWARP for Storage Performance Dataplane Kit (SPDK). This solution enables new efficiencies and performance by offloading the compute resources from the data path leveraging NVMe 1.2 Controller Memory Buffer (CMB) technology.

Driven by the performance demands of NVMe, high performance, low latency networking is a fundamental requirement for a fabric to scale out. Ethernet-based RDMA fabrics with their exceptional low latency and offload capabilities have become a popular choice for NVMe-oF. Cavium FastLinQ 10/25/40/50/100GbE Ethernet adapters support Universal RDMA (RoCE, RoCEv2 and iWARP) and deliver the ultimate choice to customers for scaling out NVMe over a general-purpose Ethernet fabric. In addition, Cavium LiquidIO line of SmartNIC solutions are key enablers for delivering JBOF/FBOF management and AFA storage services like full IPSEC offloads, compression and deduplication.

The ThunderX2 product family is Cavium’s second-generation, 64-bit ARMv8-A server processor SoCs for the data center, cloud, storage and high-performance computing applications. The family integrates high-performance custom cores along with outstanding I/O connectivity, memory bandwidth and capacity.

Delivering high single core performance and multiple PCIe lanes per socket, Cavium ThunderX2 SoCs enable highly optimized JBOF/FBOF configurations.

The XPliant family of Ethernet switches improves compute and storage application performance in cloud, enterprise, and service provider data centers by offering a variety of features. The XPliant Programmable Architecture enables switching solutions that can be optimized for specific network architectures, including storage networks. Integrated end-to-end network performance management via the Packet Trakker instrumentation tool suite and large fully shared packet buffer guarantee robust lossless performance required for scaling out NVMe-oF.

“NVMe technology is rapidly changing the way cloud and enterprise data centers connect to shared storage, but building NVMe-oF solutions require industrywide technology integrations,” said Christopher Moezzi, vice president of marketing, Ethernet Server Adapter Group, Cavium. “Cavium FastLinQ Universal RDMA technology combined with the broader Cavium and Microsemi portfolio delivers a high performance and validated reference platform for JBOF/FBOF implementations, reducing risk and time to market for joint storage customers.”

“We are excited to collaborate with Cavium as part of Microsemi’s Accelerate Ecosystem. Microsemi’s leadership in NVMe connectivity and peer-to-peer technology reduces time to market and time to revenue for our joint customers in creating NVMe-oF solutions,” said Andrew Dieckmann vice president of marketing for Microsemi. “The scale-out capabilities of Microsemi’s Switchtec PCIe switches and P2P data plane offload combined with Cavium’s unique concurrent RoCE and iWARP NICs provides compelling flexibility, scale and performance for enterprise data center and cloud applications.”

“Marvell’s NVMe cloud and enterprise SSD controller data center customers are seeking to significantly scale and disaggregate their storage solutions without compromising latency, performance or reliability,” said Nigel Alvares, vice president of SSD and Data Center Storage Solutions at Marvell. “Cavium’s unique Universal RDMA capabilities built into the FastLinQ 41000 Series Ethernet NICs, combined with its integration across the broad Cavium portfolio, are well-positioned to meet the requirements of next-generation data centers while offering best-in-class total cost of ownership.”

Cavium also announced that it is collaborating with Hewlett Packard Enterprise (HPE), to bring hyperscale-inspired designs based on Open Compute Project into mainstream enterprise IT.

The introduction of Cavium FastLinQ 41000 Series 10/25GbE NICs in OCP 2.0 form factor for HPE’s Cloudline Servers and broad collaboration on driving the emerging OCP NIC 3.0 standard will enable mainstream delivery of the most efficient designs for scalable computing to Cavium and HPE’s enterprise and cloud service provider customers.

Tier-1 hyperscale buyers define leading-edge operational efficiency and have the skill set along with the purchasing power to request or design custom and semi-custom hardware, which enables operational cost savings and performance benefits to enable their businesses to run more efficiently. Enterprise IT organizations and cloud service providers outside the hyperscale category often lack the scale and skill set for custom hardware design but are increasingly under more pressure to improve efficiency and provide competitive differentiation for their business.

As a result, minimizing capital expenses and operating expenses is a key priority for Enterprise IT as they seek disruptive innovations or new business models to achieve their goals. These organizations are closely tracking hyperscale buyers’ actions and are looking for ways to leverage hyperscale best practices in their own IT environments. To optimize visibility, improve efficiencies, and achieve collaborative design opportunities with the Enterprise datacenter supply chain, CIO and CTOs are demanding technologies that are considered “open.”

Purpose-built to accelerate and simplify data center networking, Cavium FastLinQ Ethernet OCP NIC technology in HPE Cloudline servers delivers universal RDMA network adapter that offers customers a choice of RDMA technology and investment protection with concurrent support for RoCE, RoCEv2 and iWARP; network virtualization offloads that accelerate for network virtualization by offloading protocol processing for VxLAN, NVGRE, GRE and GENEVE, enabling customers to build and scale virtualized networks without impacting network performance.

The technology also optimizes infrastructure costs and increases virtual machine density by leveraging built-in technologies like SR-IOV and NIC Partitioning (NPAR) that deliver acceleration and QoS for workloads and infrastructure traffic. Its Network Function Virtualization (NFV) enables Telcos and NFV application vendors to seamlessly deploy, manage and accelerate the most demanding NFV workloads by delivering leading small packet performance and integration with DPDK and OpenStack.

Cavium and HPE were both quick to recognize the value that the OCP standard can bring to customers and have been committed to OCP’s vision since its inception. HPE Cloudline uses standardized and custom servers that leverage OCP mezzanine adapters, such as the Cavium FastLinQ 41000 Series 10/25GbE NICs.

Currently available from HPE in OCP 2.0 form factors, Cavium FastLinQ adapters enable easier adaptability, scalability, and integration with support for key features such as network virtualization offload and Universal RDMA. Further extending their leadership in OCP, Cavium and HPE are collaborating to jointly drive, evangelize and embrace the Open Compute Project NIC 3.0 standard across the server and networking community worldwide.

“Cavium has a decade-old relationship with Hewlett Packard Enterprise (HPE) and provides differentiated I/O technology for HPE servers, storage and networking,” said Rajneesh Gaur, Vice President and General Manager of Ethernet Adapter Group, Cavium. “As technology pioneers and innovators of open standards, I am proud to announce that Cavium and HPE are aligning their strategy and investments to enable Enterprise and Cloud IT to leverage the economics of scale by driving the OCP NIC 3.0 standard.”

“HPE is committed to decreasing data center TCO while improving performance and availability by enabling an Open Infrastructure,” said Kara Long, Vice President and General Manager for Cloudline Portfolio at HPE. “With the introduction of Cavium FastLinQ OCP 2.0 NICs in Cloudline servers and collaboration with Cavium on the OCP NIC 3.0 standard development, joint customers will have the ability to transform their infrastructure for modern workloads and cloud initiatives while optimizing infrastructure costs.”

Leave a Reply

WWPI – Covering the best in IT since 1980