What is the Open Compute Project?

Why 400+ companies are collaborating on open hardware for data centers.

The Open Compute Project is a global collaborative community focused on advancing data center hardware. Originally started by Facebook and a few partners 13 years ago, it has grown to include over 400 companies and thousands of engineers working together to tackle some of the biggest challenges in modern computing.

While the project leverages AI internally, its primary focus is on building infrastructure that can support rapidly growing AI workloads. From GPUs to liquid cooling technology, the Open Compute Project is driving innovation to ensure data centers can efficiently handle the increasing demands of AI-driven computing.

Video transcript

What is the Open Compute Project?

Open Compute Project is a global collaborative community focused on innovating data center hardware. We started by Facebook then 13 years ago and a handful of other collaborators to build the world’s most efficient data center. And today it’s grown into a membership of over 400 companies and thousands of engineers working together to solve data center’s most challenging problems.

We use it internally, our teams are small and we’ve seen our community leveraging AI, but we’re really building for AI. We’re serving these new compute loads that are using GPUs, liquid cooling technology; a lot of changes happening in the data center and computational infrastructure. So we’re really in service to the AI a little bit, than leveraging.

More from We Love Open Source

• Why you should learn assembly language
• Why Kubernetes is essential for AI and open source
• Why open hardware is essential to scaling AI workloads efficiently
• Exploring PyTorch and open source AI tools with Sahdev Zala
• Get started with gokrazy: Simple self-hosting on Raspberry Pi