Transforming Compute for Next-generation Infrastructure

By Chris Bergey, SVP and GM, Infrastructure Line of Business, Arm

News Highlights:

• The Arm Neoverse V1 platform is first in a new computing tier for Arm, and the first-ever Arm designed core to support Scalable Vector Extension (SVE), delivering 50% more performance for demanding HPC and ML workloads
• The Arm Neoverse N2 platform is the first based on Armv9, delivering 40% more performance for diverse workloads
• Arm Neoverse CMN-700 is the industry’s most advanced mesh interconnect to unleash the performance and performance/watt benefits of Neoverse V1 and N2 platforms

The demands of data center workloads and internet traffic are growing exponentially, and new solutions are needed to keep up with these demands while reducing the current and anticipated growth of power consumption. But the variety of workloads and applications being run today means the traditional one-size-fits all approach to computing is not the answer. The industry demands flexibility; design freedom to achieve the right level of compute for the right application.

As Moore’s Law comes to an end, solution providers are seeking specialized processing. Enabling specialized processing has been a focal point since the inception of our Neoverse line of platforms, and we expect these latest additions to accelerate this trend.

Back in September when we introduced these new Neoverse N2 and Neoverse V1 platforms to our roadmap, it was a testament to the groundswell of Neoverse interest and adoption across our growing ecosystem. Today we’re pleased to share some of the new features of these platforms, and the growing list of Arm partners bringing the superior performance, performance per watt and TCO benefits of Neoverse to infrastructure markets.

• Marvell revealed its OCTEON family of networking solutions based on Neoverse N2 will begin sampling by end of 2021, providing a 3x performance uplift over previous generation OCTEON solutions
• Ministry of Electronics and Information Technology (MeitY) has announced it will join SiPearl and ETRI in licensing Neoverse V1 for its national exascale HPC project
• Oracle plans to adopt Ampere Altra CPUs for Oracle Cloud Infrastructure, as the price/performance leader across a wide range of workloads
• Arm-powered AWS Graviton2 continues to rapidly expand its EC2 footprint with steady growth and regional expansion
• Alibaba Cloud just tested the upcoming Alibaba Cloud ECS Arm instances, showing off impressive results on the SPECjbb benchmark and showing improved performance of the DragonWell JDK on Arm by 50%
• Tencent is making investments in both hardware testing and on software enablement that will allow them to adopt Neoverse technology for cloud applications

These partners are taking full advantage of what is under the hood of Neoverse platforms. This is just the tip of the iceberg for both infrastructure workload benefits and on how our partners plan to implement and take Neoverse IP to market.

Blazing a trail for the new era of computing

As we look to the next era of computing, we want to change how the industry thinks about deploying infrastructure. Innovators shouldn’t have to choose between performance or power efficiency, and Neoverse helps deliver a solution that brings the best of both worlds to enable a range of cloud-to-edge use cases.

Neoverse V1 and Neoverse N2 platforms are intended to lead the way to this next era. Let’s dig into what partners can expect from each.

Neoverse V1: A revolution in high-performance computing

Delivering a massive 50%¹ uplift, 1.8x improvement for a range of vector workloads and 4x improvement for machine learning workloads over N1, Neoverse V1 is the first in a new performance-first computing tier for Arm. Neoverse V1 gives our silicon partners the flexibility to build compute for applications more reliant on CPU performance and bandwidth while providing SoC design flexibility.

With the performance-first mindset, the design philosophy behind Neoverse V1 was to build the widest-microarchitecture Arm has ever produced to accommodate more instructions in flight in support of markets like high performance and exascale computing. The wide and deep architecture with the addition of scalable vector extensions (SVE), gives Neoverse V1 the lead in per-core performance, code longevity with SVE, and provides SoC designers implementation flexibility. You can see the benefits of some of these design elements in SiPearl and ETRI’s HPC SoCs and we think this is the direction HPC compute is heading.

For more details on the technical features of the Neoverse V1 platform, read our blog here.

Neoverse N2: Market leading cloud-to-edge performance

A few weeks ago, Arm introduced the Armv9 architecture to address global demand for ubiquitous specialized processing. The Neoverse N2 platform is paving the way for infrastructure cores and is the first based on the Armv9 architecture with improvements to security, power efficiency and performance.

Delivering 40%¹ higher single-threaded performance compared to N1, Neoverse N2 still retains the same level of power and area efficiency as Neoverse N1. The scalability of Neoverse N2 extends from high-throughput computing, such as in hyperscale cloud where we see 1.3x improvement on NGINX over N1 and down to power and space constrained edge and 5G use-cases which deliver 1.2x faster DPDK packet processing over N1.

The Neoverse N2 platform delivers superior performance per-thread, and industry-leading performance-per-watt driving a reduced TCO for users. Neoverse N2 is the first platform to feature SVE2, an Armv9 feature that drives a significant uplift in cloud-to-edge performance efficiency. For a broader set of use cases like machine learning, digital signal processing, multimedia and 5G systems, SVE2 not only brings performance but also the ease of programming and portability benefit of SVE.

For more details on the technical features of the Neoverse N2 platform, read our blog here.

Building heterogeneous SoCs

We are eager to see how our partners will take advantage of the performance and scalability capabilities of these two new platforms, but we couldn’t call them “platforms” without Arm system IP. Today we are also unveiling CMN-700, a key element for constructing high-performance Neoverse V1 and Neoverse N2-based SoCs.

Building on the success of CMN-600, CMN-700 provides a step function increase in performance on every vector, from core counts and cache sizes to the number and types of memory and IO devices you can attach.

CMN-700 enables next-generation use cases for multi-chip, memory expansions and accelerators. Through our continued investment in CCIX and CXL, we’re providing more customization options and enabling partner solutions with fast fabric and high core count scalability. This will open doors to expand beyond traditional silicon limits and allow for more flexibility for tightly coupled heterogeneous compute.

You can read more about the new CMN-700 features and capabilities in each of the Neoverse V1 and Neoverse N2 technical blogs.

Liberty to innovate

There is a new wave of innovation coming from across the Arm ecosystem, and the two new products we’ve delivered today strengthen our commitment to the Arm Neoverse roadmap. By achieving higher than promised generational performance gains and elevating the design flexibility in our platforms, we are enabling partners with the freedom to build, design, and create solutions that will enable the next-generation infrastructure.

Additional Resources

• Visit the Neoverse V1 and Neoverse N2 product pages for additional details
• Learn about CMN-700
• Listen to The Future of Infrastructure podcast on Arm BluePrint
  podcast
• The how and why of estimating SPEC CPU performance for Arm Neoverse
• Breaking down Arm Neoverse performance leadership

2:41

Supplemental Quote Sheet

Cloudflare

“Cloudflare supports a network of over 100 countries, more than 200 cities and growing. So strong CPU performance and power efficiency are critical to Cloudflare being able to scale our network services in a cost-friendly manner. Modern Arm Neoverse-based CPUs are demonstrating industry-leading performance throughput and power efficiency, and we like the direction Arm is heading with the addition of their Neoverse V1 and N2 platforms.” – Nitin Rao, Senior Vice President of Global Infrastructure at Cloudflare.

Electronics and Telecommunications Research Institute (ETRI)

“For over fourty years ETRI has been contributing to electronics and computing leadership on behalf of the South Korea government and its citizens. In pursuit of advancing ABCA – Artificial Intelligence, Big Data, Cloud Computing and IoT – ETRI is developing the K-AB21 ‘artificial brain’ CPU based in part on the Arm Neoverse V1 core. Chosen for its high-performance and low power consumption, the Neoverse V1 will help us achieve our goal of 2.5x performance with 60% power reduction over alternative HPC and AI accelerator technologies.” – Dr. Youngsu Kwon, Director of the AI Processor Research Department, ETRI

Fujitsu

“Fujitsu and Arm worked together closely on the development of SVE for the HPC market. This ultimately led to the first commercial deployment of SVE in our A64FX CPU, which now powers the world’s leading supercomputer, Fugaku. We hope that the support of the Arm Neoverse V1 core will drive further development of SVE-optimized HPC codes to grow the overall HPC market. We anticipate that Arm and Fujitsu will continue to improve each other’s presence in the global HPC market by making the best use of the strengths of their respective products.” – Naoki Shinjo, Corporate Executive Officer, SVP, Deputy Head of Future Society and Technology Unit, Fujitsu

Marvell

“Arm’s Neoverse N2 offers the industry’s top computing power per watt, delivering three times the compute performance and four times the SPECint per watt of the current generation. Our next-generation OCTEON DPU family will leverage Arm’s Neoverse N2 cores to power critical infrastructure applications from 5G to storage to signal processing and security.” – Raj Singh, Executive Vice President of the Processors Business Group at Marvell

Ministry of Electronics and Information Technology (MeitY)

“We are working towards complete indigenization of Supercomputing technology under National Supercomputing Mission (NSM) with support from Ministry of Electronics and Information Technology (MeitY) and Department of Science and Technology (DST). Our next goal under NSM is to establish an Exascale computing capability within India, which will be based on our indigenously designed HPC Processor SoC. For this, we have chosen the Arm Neoverse V1 platform which combines high-end CPU and vector performance with strong power efficiency, giving our SoC and System designers the freedom to optimize CPU, memory and IO capabilities for the unique requirements of Exascale computing.” – Dr. Hemant Darbari, Director General, Centre for Development of Advanced Computing (C-DAC)

NVIDIA

“NVIDIA has been at the forefront of a massive change sweeping through the datacenter market, with the advent and adoption of GPU-driven AI & Machine Learning techniques applied to challenges across all industries. We see a similar transformation taking place across the breadth of computing infrastructure, from the cloud to the edge, and we believe Arm is both a catalyst for and a solution to this change. Customer adoption of specialized silicon has never been greater, and both NVIDIA and Arm are at the center of this trend.” – Ian Buck, vice president and general manager of Accelerated Computing at NVIDIA

Oracle

“We believe Arm is going to be everywhere – from edge to the cloud. We are seeing N1 based processors deliver consistent performance, scalability and security that customers want from Cloud infrastructure. Partnering with Ampere Computing and leading ISVs, Oracle is making Arm server-side development a first-class, easy and cost effective solution.” – Bev Crair, Senior Vice President, Oracle Cloud Infrastructure Compute.

Red Hat

“The history of collaboration between Red Hat and Arm spans from architecture to standards to ecosystem support. Red Hat Enterprise Linux, the world’s leading enterprise Linux platform, offers architectural choice to our customers and already supports a broad range of processors, running from cloud to edge, that are based on the building blocks of Arm Neoverse. We look forward to extending this support to future Neoverse V-series and N-series processor designs based on the Armv9 architecture, which furthers support for zero trust architectures via the ‘realms’ capability.” – Jon Masters, Distinguished Engineer at Red Hat

SiPearl

“SiPearl is on a mission to design and bring to market CPU technology that will power Europe’s first exascale supercomputer. SiPearl has selected the Arm Neoverse V1 core as the general-purpose processing element of our design, due to it’s high performance and energy efficiency and Arm support for Scalable Vector Extensions (SVE). We are excited about the potential of Arm-based computing for HPC and ML-accelerated applications and look forward to contributing to the scientific community.” – Philippe Notton, CEO and Founder, SiPearl

SUSE

“SUSE has played a significant and active role in the Arm ecosystem. From the early days of enterprise-grade Linux on 64-bit Arm, SUSE Linux Enterprise Server on Arm is a secure, reliable and adaptable operating system that is widely available on Arm infrastructure, from Raspberry Pi to high performance computing environments. Now combined with Rancher, SUSE offers a wide range of use cases on Arm, including for containerized workloads, from data centers to the edge. SUSE looks forward to ongoing collaboration with Arm to help enterprises reduce time to value with optimized Arm architecture solutions—including on the next generation of Arm Neoverse processor cores—allowing customers to innovate everywhere.” – Thomas Di Giacomo, chief technology and product officer at SUSE

Performance and Benchmark Disclaimer

¹Uplift is based on IPC assuming constant process and frequency.

This benchmark presentation made by Arm Ltd and its subsidiaries (Arm) contains forward-looking statements and information. The information contained herein is therefore provided by Arm on an “as-is” basis without warranty or liability of any kind. While Arm has made every attempt to ensure that the information contained in the benchmark presentation is accurate and reliable at the time of its publication, it cannot accept responsibility for any errors, omissions or inaccuracies or for the results obtained from the use of such information and should be used for guidance purposes only and is not intended to replace discussions with a duly appointed representative of Arm. Any results or comparisons shown are for general information purposes only and any particular data or analysis should not be interpreted as demonstrating a cause and effect relationship. Comparable performance on any performance indicator does not guarantee comparable performance on any other performance indicator. Any forward-looking statements involve known and unknown risks, uncertainties and other factors which may cause Arm’s stated results and performance to be materially different from any future results or performance expressed or implied by the forward-looking statements. Arm does not undertake any obligation to revise or update any forward-looking statements to reflect any event or circumstance that may arise after the date of this benchmark presentation and Arm reserves the right to revise our product offerings at any time for any reason without notice.Any third-party statements included in the presentation are not made by Arm, but instead by such third parties themselves and Arm does not have any responsibility in connection therewith.