An Initiative to Streamline Software Development on Arm with Free Cloud Resources

The Works on Arm initiative enables developers to build, test, and optimize projects on the Arm64 architecture by providing free-of-cost access to Arm based developer platforms, cloud instances, and CI/CD environments. Arm has partnered with cloud platform providers to make Neoverse-based compute available for cloud-to-edge solutions, while offering choice and flexibility.

 

Developer Offerings:

The GitHub Arm-based hosted runners, seamlessly integrated with GitHub Actions, empower native application development and testing on Arm Neoverse platforms. Preloaded with essential tools and a cloud developer image, these versatile runners efficiently support diverse Linux and Windows application development across cloud, edge, and automotive use cases.

Amazon Web Services (AWS) provides a highly reliable, scalable, low-cost infrastructure platform in the cloud that powers hundreds of thousands of businesses in 190 countries around the world.

Equinix Metal provides globally-available bare metal “as-a-service” infrastructure that can be deployed and interconnected in minutes.

Whether your business is early in its journey or well on its way to digital transformation, Google Cloud helps you solve your toughest challenges.

Microsoft Azure cloud platform consists of more than 200 products and cloud services designed to help you bring new solutions to life — to solve today’s challenges and create the future. Build, run, and manage applications across multiple clouds, on-premises, and at the edge, with the tools and frameworks of your choice.

miniNodes is an Arm-centric innovation, prototyping, research, and development, consulting, and experimentation organization.

The Open Source Lab (OSL) is a nonprofit organization contributing to the advancement of open source software and technologies.

Oracle Cloud Infrastructure (OCI) combines the elasticity and economics of a public cloud with the granular control, security, and predictability of on-premises infrastructure.

Tencent Cloud launched the first new generation of Arm based standard SR1 Cloud Virtual Machines (CVM). These CVMs provide reliable performance, lower power consumption, and lower costs.

Works on Arm Testimonials

"Adoptium produces Eclipse Temurin binaries on both 32 and 64 bit Arm platforms. Most of these systems are from the Works on Arm initiative and are currently supporting four parallel versions of OpenJDK builds, performing testing across a broad range of Linux distributions with ease. The additional systems provided by the Works on Arm initiative resulted in our Linux on Arm build and test cycle being faster than many other platforms." Stewart Addison, Eclipse Adoptium Project
"Works on Arm is helping Debian packages to be built on Arm through Continuous Integration (CI) and to execute automated tests required for a stable release. More than 3 million CI jobs have run on these platforms as of October 2022. Debian thanks Works on Arm for this wonderful support." Santiago Ruano Rincón, Debian Project
"The hardware provided by Works on Arm has been instrumental in providing resources to the open-source DynamoRIO project for both automated testing and interactive development for the AArch64 platform. AArch64 is a key target for the DynamoRIO community. Without this hardware we would be limited to emulation which has many shortcomings and restrictions including being substantially more difficult to debug and being unable to successfully run many of our tests." DynamoRio Project
"Arm-based platforms provided by Works on Arm continues to support OCaml in processing 1000+ Aarch64 jobs per day in our Continuous integration cluster. This OCluster is used by OCaml developers to test their packages at scale to release quality software." Anil Madhavapeddy, OCaml Project
"Through the Works on Arm initiative the upstream CI fleet of the Python project has been expanded to provide coverage for the Arm architecture on various Linux Operating Systems. We have been able to catch early regressions leading to fixing not only bugs in the core Python language, but even in the Linux kernel and various toolchains, uncovered through Python's extensive test suite. This has helped establish Arm as one of the first-class architectures within the Python language ecosystem." Charalampos Stratakis, Python Language
"AlmaLinux made it a point to support and provide a first-class experience on Arm right out of the gate with our distribution. We've also made it a point to enable others to be able to use Arm as well. The ELRepo project is a project which builds various Linux kernels and modules for the Enterprise Linux ecosystem, and we enabled resources for them to be able to support Arm. We are sincerely thankful to Arm and their partners for all the support they've provided us, whether it be hardware or technical resources. Arm, and Arm developers have become a vital part of our community." Jack Aboutboul, Director and Community Leader, AlmaLinux
"The Node.js project provides prebuilt binaries of the open-source server-side JavaScript runtime on a variety of platforms. Our continuous integration builds and tests Arm v8 and Node.js on a regular basis including on pull requests opened by our contributors. The Works on Arm initiative provides us with the machines required for those builds on Arm. In addition, these systems support running many Jenkins jobs in parallel and we have used the Ampere Altra systems to support building and testing 32-bit armv7l distributions of Node.js in containers on those hosts. This has reduced our dependency on the less reliable single board computers including a number of Raspberry Pis which we were previously using for that work." NodeJS Project
"The Julia community has worked with Works on Arm to greatly improve its aarch64 platform support for Julia. This made it possible for the Arm platform to become a supported platform on Linux. Over the past year, our Arm binaries have seen over 100,000 downloads (only from julialang.org), and support over 15,000 users." Viral Shah, Co-creator of Julia and CEO, Julia Computing

AWS

AWS Graviton processors, based on Arm Neoverse 64-bit cores, are designed by AWS to deliver the best price performance for your cloud workloads running in Amazon EC2.

 

AWS Graviton3 processors are the latest in the AWS Graviton processor family. They provide up to 25% better compute performance, up to 2x higher floating-point performance, and up to 2x faster cryptographic workload performance compared to AWS Graviton2 processors.

 

All new and existing AWS customers can try the t4g.micro instances free until December 31, 2022. During the free-trial period, developers who run a t4g.micro instance will automatically get 750 hours per month deducted from their monthly bill.

Equinix

Arm and Equinix have a long-standing partnership to make powerful Armv8 bare metal infrastructure — including latest generation of Ampere systems — available to the open-source software ecosystem to build, test, and optimize for Arm64 architecture. The Ampere Altra are available in single socket (80 cores, 256GB RAM, 1TB NVMe U.2 SSD) and dual socket (160 cores, 512GB RAM, 2TB NVMe U.2 SSD) configurations where Ampere eMags are available in a single socket (32 cores, 128GB RAM, 480GB SSD) configuration.

 

Equinix, as part of the Work on Arm initiative, provides free access to state-of-the-art computing resources for software developers, supporting a wide variety of projects with a special focus on building systems, languages, and cloud native applications. A common use case for these resources is for public CI and setting up self-hosted runners. The on-demand infrastructure resource is managed by Equinix Metal as part of its commitment to the Arm ecosystem.

Google Cloud

Powered by Ampere® Altra® Arm-based processors, Google Cloud VMs deliver exceptional single-threaded performance at a compelling price. VMs come in multiple predefined VM shapes, with up to 48 vCPUs per VM, and 4GB of memory per vCPU. They offer up to 32 Gbps networking bandwidth and a wide range of network-attached storage options, making Google Cloud VMs ideal for scale-out workloads, including web servers, containerized microservices, data-logging processing, media transcoding, and large-scale Java applications.

Microsoft Azure

Run the apps your customers want on Azure Virtual Machines featuring Ampere® Altra® Arm-based processors. Develop and deploy a variety of workloads such as web and application servers, open-source databases, Java and .Net applications, gaming, and media servers, and more.

Engineered to efficiently run scale-out cloud-native workloads, the Dps v5 and Dpls v5 VM series offer excellent price-performance. Additionally, memory optimized Eps v5 VM series are designed to meet the requirements of memory-intensive workloads such as open-source databases, in-memory caching applications, and data analytics engines. All the Arm-based Azure VMs can be included in Kubernetes clusters managed using Azure Kubernetes Service (AKS).

Get started now with pay-as-you-go pricing. There’s no upfront commitment—cancel anytime. Or try Azure free.

miniNodes

miniNodes provides thought leadership and ecosystem building by exploring topics such as clustered computing on small form-factor Arm devices, building the world’s first IoT, edge, and cloud proof-of-concept running entirely on Arm platforms, demonstrating the energy efficiency of Arm Servers, and advocating for standards to help ease the burden of booting Arm devices.

 

In addition, miniNodes has built and maintained a small hosting infrastructure for several years that provides low-cost, low-power Arm devices on a public IPv4 internet connection for developers and enthusiasts to run small servers. Primarily consisting of Raspberry Pis and Nvidia Jetson Nanos, this public-facing service has allowed miniNodes to engage with the community, gather feedback from Arm developers, and help improve the developer experience.

OSL (Open Source Lab)

The nonprofit Oregon State University Open Source Lab (OSUOSL) is accelerating the growth of key open source software projects by distributing software globally to millions of users, thus promoting the culture of open source and increasing developer productivity.

 

OSL provides Ampere Computing’s Arm Neoverse-based eMAG servers to develop and test open source software for the Arm64 architecture. The cluster is based on an Open Stack environment and offers Arm64 instances running on KVM, accessible via OpenStack’s GUI and API interface. These Arm64 server resources are ready and open to the developer community upon request.

Oracle

Oracle Cloud Infrastructure (OCI) is a platform of cloud services that enable developers to build and run a wide range of applications in a highly available, consistently high-performance environment. OCI is designed to help companies run their entire application portfolio in the cloud, especially mission-critical workloads.

 

The Ampere A1 Compute platform brings additional diversity and choice to cloud computing and in partnership with Ampere Computing, the Altra Arm processors are available on OCI. Oracle is the only cloud provider offering Arm-based compute instances at only one cent per core hour with flexible sizing from 1 to 80 OCPUs and 1 to 64 GB of memory per core. The Ampere A1 compute platform provides deterministic performance, linear scalability, and a secure architecture with excellent price-performance.

Tencent

Tencent cloud's first new instance of Arm architecture is powered with Ampere Altra processors, with a frequency of 2.8GHz, and is based on the high-performance Arm Neoverse N1 cores. Fully applicable to container applications, websites, test development and other general computing scenarios. Arm cloud servers will play its special architectural advantages in Android native scenarios such as cloud mobile phones, Android simulation tests, and embedded development to meet more application scenarios. Tencent's self-developed Tencent Kona JDK provides Java runtime support for new instances, has passed complete functional testing, and users can choose to run Java applications in Kota JKK 8 or 11.