Google is bringing its open-source platform for machine learning to the enterprise. According to the company, enterprises that want to use TensorFlow struggle because of the higher demands and expectations they have for running ML workloads.
TensorFlow Enterprise will feature enterprise-grade support, cloud scale performance and managed services. “Together, these services and products can accelerate your software development and ensure the reliability of your AI applications,” Craig Wiley, director of product management for the cloud AI platform at Google, wrote in a post.
Enterprise-grade support will include long-term version support for companies who invest heavily in previous TensorFlow versions. Certain versions will get security patches and select bug fixes for 4 years and be supported on Google Cloud. In addition, support will feature engineer-to-engineer assistance from Google Cloud and TensorFlow teams.
Cloud scale performance will include deep learning VMs (which are generally available) and deep learning containers (which are in beta) to help enterprises get started and scale.
The solution will also provide the same experience and enterprise-grade features as Google Cloud’s managed services.
More information is available here.