Google Announces Open Source ML Compiler Project, OpenXLA

(Pasko Maksim/Shutterstock)

At its Next ‘22 event, Google Cloud announced the launch of an open source machine learning compiler ecosystem, OpenXLA.

OpenXLA is a community-led and open source ecosystem of ML compiler and infrastructure projects co-developed by Google and other AI/ML developers including AMD, Arm, Meta, NVIDIA, AWS, Intel, and Apple.

Incompatibilities between frameworks and hardware can impede ML development, and OpenXLA aims to address this by giving flexibility to developers when it comes to the frameworks and hardware they choose for ML projects.

According to the OpenXLA project charter, the objective of the project is to enable efficient lowering, optimization, and deployment of ML models from most major frameworks, such as PyTorch and TensorFlow, to any hardware backend (notably CPUs, GPUs, and ML ASICs) through collaborative work with major ML frameworks and hardware vendors.

In a blog post, Sachin Gupta, Google VP and GM of infrastructure, explains that the first objective for the new community project will be a collaborative evolution of the XLA compiler, a compiler that was developed to streamline modeling in TensorFlow through speeding up the training process and reducing overall memory consumption. The compiler is now being decoupled from TensorFlow, and OpenXLA will work to build StableHLO, a portable ML compute operation set which acts as a portability layer between machine-learning frameworks and compilers.

OpenXLA’s goals are listed as the following:

• Accelerate industry collaboration around XLA and build a vibrant OSS community.
• Share and receive feedback on the technical direction for OpenXLA and ensure it meets the needs of major users and contributors.
• Set up a new XLA repository or organization with independent build/test, with infra to more easily accept PRs, and that is hardware and framework independent.
• Ensure the extraction of XLA from TensorFlow is minimally disruptive to existing users and contributors.
• Create a product identity with its own brand, website, docs, and communication channels.
• Discuss establishment of governance outside TensorFlow.

“At Google, we believe open-source software is essential to overcoming the challenges associated with inflexible strategies. And as the leading Cloud Native Computing Foundation contributor, we have over two decades of experience working with the community to turn OSS projects into accessible, transparent catalysts for technological advance. We’re committed to open ecosystems of all kinds, and this commitment extends to AI/ML—we firmly believe no single company should own AI/ML innovation,” said Gupta.

Membership in OpenXLA is open to everyone involved in developing or integrating with XLA, including representatives of ML frameworks, hardware platforms, users, and integrators. To participate, members can request an invitation to join the GitHub organization and SIG Discord, to be announced at a later date.

Related Items:

Databricks Announces Major Contributions to Flagship Open Source Projects

AMD Joins New PyTorch Foundation as Founding Member

Open Data Hub: A Meta Project for AI/ML Work