Chris Bergey, SVP and GM, Infrastructure Line of Business, Arm
May 25, 2021 -- The Arm ecosystem continues to leverage the freedom to build, design, and create solutions using Arm IP, enabling a new era of computing rooted in system innovation. The recently announced Armv9 architecture is the answer to a growing need from solution providers for specialized processing, and this is especially impactful in the infrastructure industry where a one-size-fits all approach to computing was the only option for many years.
Arm Neoverse platforms have been designed to lead the way to the next era of infrastructure computing, uniquely positioned to address and offer specialized processing that fits various workloads by delivering consistent performance, scalability, and security. The transformation in infrastructure computing is only in the beginning chapters, with initial impacts of Neoverse just beginning to be seen, with much more of the story to come as partners share plans to take Neoverse IP to market. The continued growth and strength of the Arm ecosystem is proof of the benefits of superior performance and lower power consumption Arm can provide in comparison to traditional architectures.
The power of choice in the next era of infrastructure
To date, Arm partners have shipped more than 190 billion Arm-based chips. This number represents how pervasive Arm technology is, touching so many aspects of our everyday life. From cloud-to-edge, Arm’s presence continues to grow at a rapid pace as Arm's silicon partners shipped a record 7.3 billion Arm-based chips in the last calendar quarter of 2020, equating to more than 900 chips per second and 70 millon chips a day.
This growth is even more evident with today’s news that Oracle is bringing Arm-based instances to Oracle Cloud Infrastructure (OCI) with the Ampere A1 Compute. What I am most excited about with this announcement is Arm’s ability to participate in Oracle’s approach to cloud services, which channels the company’s legacy in enterprise IT and is built to run on enterprise applications as well as cloud-native workloads. It is great to see partners like Oracle now offering Arm-based cloud solutions in markets supporting traditional enterprise workloads.
With this new cloud offering, OCI is bringing to market the industry's first 80-core Arm-based server SoC and the first cloud-native CPU for modern cloud and edge computing data centers, the Ampere Altra. At only $0.01 per core hour and flexible sizing from 1-80 CPUs, the new OCI Ampere A1 Compute cloud instances provide additional choice for developers to build the next generation of Arm-based high-performance computing applications like video encoding, scale-out web applications such as MySQL, or ML inferencing. We expect these cloud instances to offer incredible price/performance across all cloud workloads and in fact, you can already see some of these performance benefits on NGINX and when tested on applications like x264 video encoding where the new Arm-based OCI Ampere A1 cloud instances are delivering up to 98% better price/performance compared to traditional x86-based instances.
The result of strong collaboration
When we leverage the breadth of the Arm ecosystem, we are able to bring the benefits of collaboration to the developer community. This new cloud solution from OCI is the result of Oracle, Arm, and Ampere bringing together what makes each company unique so that customers can run Arm-based applications easily in the cloud.
Ampere has brought to life the Arm Neoverse technology blueprint in an 80-core Ampere Altra CPU, delivering performance, linear scalability, and a secure architecture for a broad set of workloads. Oracle has then enabled key building blocks such as Linux OSes, Oracle Container Engine for Kubernetes (OKE) service, MySQL, and Oracle Java to ease the process for developers who want to build new applications or port existing applications to Arm.
These performance, scalability, and security characteristics prized by Oracle’s cloud customers are uniquely delivered through the Arm Neoverse N1 core architecture. Each N1 core is single-threaded, giving the application full access to all core execution and cache resources. This also provides security as core memory regions are never shared across application threads. And by offering excellent performance-per-watt, Neoverse N1 implementations like Ampere Altra can pack more cores into a similar CPU power footprint as that of traditional architectures.
Keeping pace with cloud computing innovation
The development happening in edge computing, AI, and data analytics means cloud solutions are becoming table stakes in information technology as a more cost effective and efficient way to process data and run applications. To keep pace with this innovation, application developers need choices and our engagement with Oracle and Ampere is building on our efforts to bring more flexibility and choice to the next era of infrastructure.
As part of today’s announcement, OCI shared it is further enabling Arm compute developers to transition, build, and run Arm-based applications with significant price-performance benefits in the cloud, and they can do this today. Oracle is offering developers an Oracle Cloud Free Tier, a 30-day free trial to start using these latest compute processors. The Always Free tier includes 4 A1 cores and 24GB of RAM for developers to get started.
Now more than ever, the strength of the Arm ecosystem in the infrastructure market is clear, and we’re eager to see what else Arm partners have in store for bringing Neoverse IP to market. The launch of the Ampere A1 Compute instance family from Oracle is giving enterprise customers a strong incentive to try Arm Neoverse. And we’re confident the performance, scalability, and security Arm Neoverse delivers – at just $0.01 per core-hour – is the type of choice and diversity the cloud market is seeking.
Arm technology is at the heart of a computing and data revolution that is transforming the way people live and businesses operate. Our energy-efficient processor designs and software platforms have enabled advanced computing in more than 190 billion chips and our technologies securely power products from the sensor to the smartphone and the supercomputer. Together with 1,000+ technology partners we are at the forefront of designing, securing and managing all areas of AI-enhanced connected compute from the chip to the cloud.