Anyscale Platform Introduces New Breakthroughs in AI
New capabilities remove traditional friction AI teams face when building, iterating, and scaling machine learning and Python workloads.
Note: TDWI’s editors carefully choose vendor-issued press releases about new or upgraded products and services. We have edited and/or condensed this release to highlight key features but make no claims as to the accuracy of the vendor's statements.
Anyscale, the company behind the Ray open source framework for scaling machine learning or Python workloads, has announced new features of the Anyscale Platform. The new capabilities extend beyond the advantages of Ray open source to make AI/ML and Python workload development, experimentation, and scaling even easier for developers.
For fast development and iteration, the new Anyscale Workspaces environment is now available for early access. Workspaces provides a unified and seamless developer experience to scale ML workloads from a laptop to the cloud with no code changes. In a single environment, developers can now build and move workloads to production while still leveraging familiar tools.
Furthermore, for accelerated development and rapid iteration, the Anyscale Platform adds the ability to start up clusters up to 5 times faster than Ray open source, so developers can further accelerate iteration, experimentation, and deployments; job scheduling automation including autoscaling, alerting, autoretries and more; and custom cluster environments to provide organizations even more deployment and hosting flexibility.
New Capabilities and Highlights
Anyscale Workspaces: Workspaces provides a unified and seamless laptop-like development experience to build and scale ML workloads. Workspaces enable developers to continue using the tools they are familiar with, including VS Code, Jupyter, the terminal, and more, but leverage the scale and flexibility of the cloud. With a single script a developer can prepare data, tune, train, and deploy workloads at any scale.
As one team in a manufacturing conglomerate said, “Anyscale Workspaces allows me to go from development to experimenting at scale all the way to production all within the same environment. Workspace reduces context switching for us by 50 percent and integrates easily with the other tools we use.”
Fast cluster setup: Machine learning model training and tuning is inherently iterative, and each iteration often requires cluster startup, tuning, and shutdown events. Anyscale shortens iteration cycles by taking cluster startup events down to under two minutes -- up to 5 times faster than Ray open source.
Custom cluster environments: Organizations can now deploy their own custom Docker images as Anyscale cluster environments and leverage their existing CI/CD pipelines to build and manage workloads running on Anyscale and Ray. This includes launching Anyscale Workspaces, jobs, and services while leveraging their own Docker tooling and infrastructure.
Job scaling and automation: Anyscale now provides a native way to schedule jobs in addition to integrating with best-of-breed orchestration tools such as Airflow and Prefect. With Anyscale job automation and through integrations, Anyscale provides autoscaling, alerting, autoretries, and more to simplify moving workloads to production.
To learn more about Anyscale, please visit https://www.anyscale.com/.