Anaconda, a Python data science platform provider, is releasing Anaconda Enterprise 5.2, adding NVIDIA GPU-accelerated, scalable machine learning and more.
“The purpose is to help the data scientist be productive and at the same time give the IT folks the control and automation they need to be successful with large scale deployments, machine learning, and artificial intelligence,” said Mathew Lodge, SVP products and marketing, Anaconda.
The company worked with NVIDIA to build out support for running the platform at scale on cloud native infrastructure, Lodge explained.
Anaconda Enterprise is a software platform for developing, governing, and automating data science and AI pipelines from laptop to production.
Anaconda Enterprise uses cloud native approaches, including Docker and Kubernetes, to scale data science and machine learning across teams and clusters while simplifying and automating AI/ML governance and reproducibility.
“That’s the key mechanism we used to build out scalable machine learning,” Lodge said.
This means data scientists can share essential clusters of GPUs for efficiency and speed, according to Lodge.
“It’s very easy for the data scientists to run their code against a GPU, it’s a one-click deployment to do that and the IT organization doesn’t have to buy every data scientist their own GPU they can share a cluster,” Lodge said.
In addition to this new capability, job scheduling has been updated to help update models and schedule jobs at any time. Along with job scheduling, collaboration between different teams and tighter governance capabilities has been included in this latest update.
Anaconda Enterprise is the AI enablement platform that provides the foundation for AI/ML libraries and toolkits (e.g., TensorFlow, Scikit-Learn, MXNet, PyTorch and XGBoost), empowering organizations to deploy and manage them quickly and easily.
Anaconda Enterprise integrates directly with the organization’s authentication, source code control, and data lakes and ensures end-to-end governance and control.
For more information about this update, visit www.anaconda.com.