As AI evolves, the value of deep learning continues to grow, but implementing it can be a complex, risky, and time-consuming process. Intel is looking to change that with its new open-source platform for distributed deep learning in Kubernetes, Nauta.
Nauta provides a distributed computing environment for training deep learning models on systems running on Intel’s Xeon Scalable chips. Nauta can also be used to define and schedule deep learning experiments and check the status or results of those experiments.
According to Intel, developers will be able to use existing data sets, proprietary data, or download data. They can also create public and private folders, making collaboration between teams easier. Job inputs and outputs can also be shared among team members, Intel explained.
“Nauta gives users the ability to leverage shared best practices from seasoned machine learning developers and operators without sacrificing flexibility. At every level of abstraction, developers still have the opportunity to fall back to Kubernetes and use primitives directly. Nauta gives newcomers to Kubernetes the ability to experiment — while maintaining guard rails,” Intel wrote in a post.