Ampere Computing

Ampere is a semiconductor design company for a new era, leading the future of computing with an innovative approach to CPU design focused on high-performance, energy efficient, sustainable cloud computing. As a pioneer in the new frontier of energy efficient high-performance computing, Ampere is in a leading position driving sustainable computing for the Cloud, AI inferencing, and edge applications.

Ampere Computing brought to market the first credible, merchant Arm-based silicon for general compute. Azure, Google Cloud (GCP), Oracle Cloud Infrastructure (OCI) and others have adopted Ampere's CPUs to deliver Arm-based cloud instances. The one notable exception is AWS, which developed its own Arm chip, called Graviton.

Technology startup Ampere launched AmpereOne microprocessor for data centers, aiming to win a piece of that increasingly competitive market with chips designed to draw less electricity. AmpereOne microprocessors have 192 computing cores. AmpereOnes double the number of cores in rival products from AMD, and triple what Intel offers in its Sapphire Rapids chips.

 Ampere’s cloud-native CPUs for AI.

• Single-threaded design. Because each Ampere core is single-threaded, performance is essentially guaranteed - no pipeline stalls, caching contention issues, or memory bandwidth challenges. While this may sound somewhat trivial to a business user, each has a measurable impact on a workload, such as inference that has real-time or near-real-time performance requirements.
• High core counts. Up to 192 single-threaded cores means incredible scalability and the best in parallelism in theory. In inference, handling so many simultaneous requests means faster results. And to a business, this means happier customers.
• Cost-effectiveness. Defining the best inference platform is different for each organization. Performance per dollar or performance per instance cost (as inference is a popular cloud workload) is a relevant measure. And this is another area where Ampere has demonstrated leadership.

Ampere Computing is a modern semiconductor company delivering the first cloud native processors built for the sustainable cloud. Ampere designed the industry’s first Cloud Native Processors based on the ARM Aarch64 instruction set architecture (ISA). The Ampere Altra Family delivers up to 128 single threaded, power-optimized computing cores per CPU, each with large dedicated caches for superior performance of cloud native workloads.

Optimized for Cloud

By providing a new level of predictable scale-out performance and efficiency, Ampere is working with leading cloud service providers, OEMs, and an expansive partner ecosystem to deliver Cloud Native Processors that can handle the compute demands of today and tomorrow.

Optimized for AI

Ampere® Altra® Max processors deliver exceptional performance and power efficiency for AI workloads.

Ampere Altra Max is the only cloud processor that currently supports the fp16 data format while supporting fp32, and int8 data formats.

Finally, Ampere Altra Family CPUs support leading NVIDIA GPUs for heavy workload training or inferencing.

Energy-efficiency

Climate change, government policies, and ethical imperatives are prompting organizations to drive sustainability through their supply chains and data centers.

Environmental, social, and governance (ESG) goals have taken on greater urgency with accelerated global climate change and increasingly limited natural resources.

Scalable and efficient, Cloud Native Processors reduce data center power consumption and physical space requirements—a foundation for rack level efficiency and the means of a truly sustainable future. 

Cloud Native Processors deliver better efficiency at scale than legacy x86 processors, without sacrificing performance.

Ampere and its New Generation of Cloud Datacenter Processors

In 2020, Ampere launched its new generation 80-core Ampere® Altra® processors and in 2021 extended that product family with the 128-core Ampere® Altra® Max processors. The Altra design with 80 Neoverse N1 cores was the first merchant Arm silicon on the market who really went "all-out" in terms of performance targets, aiming for the best of what AMD and Intel had to offer, ending up in a very competitive standing against the newest EPYC CPUs and leapfrogging Intel’s offerings.

In 2023, the company introduced AmpereOne™ processor family with up to 192 cores and support for 128x PCIe Gen5 lanes and 8 channels of DDR5 memory.

 While the Altra Max CPUs were based on Arm’s Neoverse N1 architecture, these new AmpereOne cores were designed in-house and built on a 5nm process. These processors remain compatible with the Arm instruction sets. By designing its own cores, Ampere was able to optimize these processors for its core user base of large cloud platforms. The new AmpereOne chips can support up to 192 cores. In the AmpereOne family, every core now also gets 2MB of L2 private cache (up from 1MB) and the company added several features that ensure that every core gets access to the same amount of memory bandwidth. There are also features like process aging monitoring, which helps these vendors better manage the lifecycle of the processor and monitor it for issues, as well as advanced droop detection and security features like memory tagging and single-key memory encryption.

Ampere® Altra® VS AmpereOne™ Processors

With the addition of AmpereOne to the portfolio, Ampere is transitioning to its own custom CPU core design while remaining ARM ISA compliant. The new product line will provide more options to customers with more cores, higher bandwidth DDR5 memory support, and support for PCIe Gen5 devices.

AmpereOne also introduces new scalability, security, and performance management features. The higher processor count translates to a higher TDP ranging from 250-400W per socket. Ampere will continue to offer the family of Ampere Altra processors, providing customers with plenty of options to suit their workloads and applications.