The release of KNIME Analytics Platform 4.2 and KNIME Server 4.11.0 brings new functionality for enterprises to solve major data science challenges, as well as features for the individual KNIME user to collaborate, access more data sources, and blend more tools.
New Features for the Enterprise
KNIME launches Integrated Deployment, Flexible Execution Deployments, Metadata Mapping, and Guided Analytics - solving four major enterprise challenges in a simple and unique way.
Integrated Deployment
KNIME Integrated Deployment moves not only the selected model, but the entire data model preparation process into production simply and automatically. This allows continuous optimization in production and, for the deployment process, saves a lot of time and eliminates the risk of errors. The capture and write workflow nodes are now available in KNIME Analytics Platform and the production workflow can be deployed on KNIME Server. New blueprint workflows for ML and Continuous Deployment are available on the KNIME Hub.
Elastic and Hybrid Execution
Elastic and Hybrid Execution leverage the enterprise infrastructure choices while covering periods of high demand, dynamically. This reduces costs by starting up special-purpose, pay-as-you-go (PAYG) executors without needing to maintain specialized hardware year round. KNIME Executor Groups and Reservation are new features in KNIME Server. KNIME Server is now available on the AWS marketplace as bring-your-own-license (BYOL), while AWS Auto Scaling can also be used on a PAYG basis with KNIME.
Metadata Mapping
The new Metadata Mapping with workflow summary uses the KNIME’s self-documenting nature and enables complete metadata mapping of all aspects of the workflow. Blueprint workflows are available on the KNIME Hub for documenting the nodes, data sources, and libraries used, as well as runtime information. These workflows can be extended to cover other aspects of good governance and lineage tracking. For example, the collection of metadata information across workflows via KNIME Server.
Guided Analytics
KNIME’s Guided Analytics applications can be customized based on reusable components and are available on the KNIME Hub. The completely reworked KNIME WebPortal, which is part of the new version of KNIME Server, is the application for the collaboration with end users.
All four of these enterprise enablers are controlled and executed using one single platform, which KNIME is able to offer thanks to its open, self documenting, and transparent architecture.
New Features for KNIME Users
The release highlights for KNIME users are: multiple spaces for sharing and a new profile on KNIME Hub, some great connector nodes, the TensorFlow2 integration and several performance improvements.
- With multiple public and private spaces on KNIME Hub, you can now organize and share your work in a more flexible way. Spaces are now accessible in the new Hub profile, which enables interacting with other users by giving kudos and seeing their work and user statistics.
- The new connector nodes enable reading and writing files in Microsoft SharePoint, accessing and manipulating data and tables in Amazon DynamoDB, as well as reading data from Salesforce and SAP (the latter requires a commercial Theobald license).
- The new TensorFlow 2 Integration enables the use of the TensorFlow 2 Python package with a few dedicated nodes.
- The performance improvements in this release include the Simple File Reader, the Joiner node, and executing Python nodes.
- In this release, we also worked on improving the user experience for integration nodes with Tableau, PowerBI, and R, and more.
More details available
Read more on our what’s new page and the complete list of new nodes, enhancements, and bug fixes in our changelog.
