Arm unveils a cloud to edge infrastructure for a world of a trillion intelligent devices

Arm has disclosed details on a dedicated roadmap and new brand of infrastructure-class IP for 5G networks and next-generation cloud to edge infrastructure.  ArmNeoversesolutions aredesigned for higher-levels of performance, security, and scalability not seen today. Innovation from microarchitecture design up through silicon, software, and systems will enable best-in-class solutions to address the diverse and evolving requirements across the entire compute spectrum.

Arm also provided a first look at its Neoverse processor IP roadmap, with early details on upcoming platforms optimized for leading-edge process nodes. The new roadmap is designed specifically for infrastructure, beginning with the launch of the “Ares” IP platform in early 2019 on 7nm and delivering staggering performance gains of 30 percent per generation through 2021. 

The Neoverse IP roadmap has been specifically architected for the unique performance, efficiency, and scalability requirements needed to keep up with changing data patterns, new workloads, and the ever-increasing demands of an infrastructure evolving to support a trillion intelligent devices.  

“Today Arm is sending disruptive shockwaves across the cloud, networking and storage world as Arm Neoverse delivers the foundation for the new global infrastructure to enable a trillion connected devices,” said Drew Henry, senior vice president and general manager, Infrastructure Line of Business, Arm. “Arm Neoverse IP will enable a broad set of our ecosystem partners to transform infrastructure from cloud to edge and push compute to where it’s most needed, store data where most appropriate, and evolve networking to securely connect this complex world.”

Henry shared his vision for the new infrastructure and the diverse range of use cases Neoverse will address, including hyperscale cloud datacenters, storage solutions, and 5G networks. Arm Neoverse is based on guiding design principles centered around:

  • World-class high performance, secure IP and architectures purpose-built for cloud native and networked workloads
  • A highly-scalable set of IP optimized for leading-edge process nodes, including “Ares” (7nm), “Zeus” (7nm+), and “Poseidon” (5nm), designed to enable systems across the infrastructure
  • A robust ecosystem empowered to build unique and diverse solutions targeting a wide range of use-cases through leveraged investment in unified software, tools, and silicon platforms 

“The modern datacenter is no longer a physical construct, but a center of data and compute residing in the cloud and on the edge. More than ever, organizations must consider distributed, connected datacenter design methods to support the data and devices coming in the 5G world,” said Patrick Moorhead, Principal Analyst, Moor Insights & Strategy. “Arm is one of those rapidly emerging in the market and with Arm Neoverse purpose-built IP, it should be well-positioned to support many of the compute spectrum needs from hyperscale to edge access.”

Designing IP and system architectures for focused markets such as server, automotive, and networking has been a key priority for Arm over the past year. In the infrastructure space specifically, Arm has already been successful as the largest architecture deployed in the global internet infrastructure with nearly 30%-unit share. This achievement highlights not only a shift in preferred architecture, but the pervasiveness of Arm-powered technologies across the entire infrastructure market.

The announcement of Neoverse underscores the continued investment Arm and the ecosystem are making to deliver more ubiquitous compute from the cloud to the edge while delivering world-class performance and efficiency for the next generation of distributed cloud to edge infrastructure.