NetApp announced the availability of Elastigroup for Microsoft Azure Spot Virtual Machines (VMs).
Paperspace announced the latest iteration of its flagship product Gradient, an MLOps platform specifically built for machine learning and deep learning development in the enterprise.
The new release introduces GradientCI, a comprehensive CI/CD engine for building, training and deploying deep learning models, without the need for DevOps or any manual configuration. Enterprise developers can now leverage Paperspace's best-in-class agile machine learning (ML) tooling and methodology across multi-cloud, on-premise, and hybrid environments.
"According to IDC, more than 85 percent of enterprises are expected to adopt a 'multi-cloud' strategy. This clearly illustrates that today's enterprises are no longer operating in a one-cloud world," said Dillon Erb, CEO & Co-founder, Paperspace." They require enhanced development capabilities with greater flexibility across multiple cloud environments to suit their specific workflow needs. We're hearing this first hand from our customers and we're responding with a new and improved release of Gradient to accelerate deep learning development across the enterprise."
Today, companies spend millions building their own internal ML pipelines that need ongoing support and maintenance. Gradient eliminates this cost by providing an "agile ML development" approach and production-ready platform to accelerate the development of their AI applications. Platform capabilities include the ability for ML teams to run Jupyter Notebooks, distributed training and hyperparameter tuning. Gradient enables developers to seamlessly deploy models, and construct sophisticated pipelines that stretch across heterogenous infrastructure — all from a single hub.
Core benefits include:
- An end-to-end platform for developing, training, and deploying models
- Continuous integration between their git repositories and Gradient
- Workflow automation with powerful pipelining and deterministic processes
- Collaborate across teams: add team members, control permissions, and increase visibility across the organization
- Leverage fully-managed and optimized Intel Nervana NNP-T accelerators or transform existing infrastructure into a powerful MLOps platform
Erb continues, "Every enterprise is also in a race to become more agile, nimble and responsive to remain competitive in today's fast-changing marketplace. Only 25 percent of a data scientist's time is spent training models — the rest of their time is spent managing infrastructure and tooling. This has created a need for machine learning developers to move faster than ever and for enterprises to decrease time to market. Paperspace is introducing this new iteration of Gradient to both provide enterprises with greater control over their models, and to reduce the model development cycle from six months down to two weeks.
"Paperspace helps inspire the next generation of AI developers; they are also poised to help unlock the power of the upcoming Intel Nervana™ Neural Network Processors," said Carlos Morales, GM, AI Software, AI Products Group at Intel. "These new chips are set to deliver groundbreaking performance and Paperspace will help companies quickly and efficiently operationalize this amazing new AI hardware."