IBM Corp. is making the Anaconda data science platform available on its Cognitive Systems platform in collaboration with Continuum Analytics.

In addition, the company announced the formation of a machine learning work group within the OpenPOWER Foundation. The group, which includes Google Inc., will work to define machine learning frameworks compatible with OpenPOWER and also IBM’s Power architecture.

Big Blue said its support for Anaconda on its Cognitive Platform will also integrate its PowerAI software for deep learning. IBM’s Cognitive Platform is based on the company’s Power8 architecture, which interfaces with Nvidia Corp.’s Tesla Pascal P100 graphics processing unit accelerators, which are designed to boost analytics workloads. The company reckons the combination will benefit from leveraging its processor and GPU acceleration for machine learning workloads. It will also help data scientists and developers to deploy deep learning applications faster and easier, the company said.

For Continuum Analytics, which develops the Anaconda platform, the goal here is to get its software deeper into enterprises while simultaneously promoting the idea of using machine learning and deep learning to create new cognitive applications. For IBM, the collaboration helps to promote its PowerAI software as a way for organizations to deploy machine learning frameworks on its Power architecture. IBM said these frameworks can be tuned for optimal performance so they can better handle commercial and hyperscale workloads.

The collaboration also means that developers will be able to access PowerAI’s libraries to deploy the enterprise version of Anaconda, said Travis Oliphant, co-founder and chief data scientist at Continuum Analytics.

The partnership comes after IBM announced in February it was working with Continuum Analytics to host Anaconda on its z/OS mainframes.

Announcing the new machine learning work group, IBM noted that several companies have signed up to the OpenPOWER Foundation recently, including open-source Linux developer Red Hat Inc., application management and developer automation firm Bitfusion, and the database company Kinetica. These companies are all using GPUs to power deep learning and other AI platforms.

Image: ddouk/pixabay