There has been a lot coming out about GitHub and Arm64 (Ampere) servers. Back in June, GitHub announced that they are opening their GitHub runners to a public beta.
And earlier this month, Github announced that their Runners for Arm64 Linux and Windows have gone GA. All great news. And now, they have released a blog post titled: Streamlining your MLOps pipeline with GitHub Actions and Arm64 runners , which goes into more detail about why you want to move to Arm64 runners:
- Optimized performance: Arm processors have been increasingly optimized for ML workloads, offering competitive performance in training and inference tasks.
- Cost efficiency: Arm64 runners are priced 37% lower than GitHub’s x64 based runners, allowing you to get more workflow runs within the same budget.
- Scalability: easily scalable within Actions, allowing you to handle growing computational demands.
And I would add that running Arm64 builds natively on Cloud Native processors removes the complexity of using emulation and a huge improvement in build performance (we have seen 2+ hour builds come down to under 10 min).
And at FOSSY, we were lucky enough to have Larissa Fortuna, PM at GitHub, give a talk titled: GitHub Actions & Arm64: Enabling the world’s software to build on ARM.
So take a moment and check out the links above and let us know your story about building natively on Ampere/Arm64 processors.