Announcing: KNIME Software Release 4.5

Over the past few years, our community has grown to include practically any kind of data worker: from Python scripters to Excel users; from data scientists to financial analysts; from chemists to HR analysts. The latest release brings a little something for everyone.

KNIME Software release features:

• A huge boost in performance for Python - Python is now as fast in KNIME as it is standalone.
• (Even) more data wrangling options - Featuring new connectors and improved integration with Excel and MongoDB.
• Workflows as services - The capability to share more KNIME data types between workflows beyond what was possible with JSON (e.g., text documents, images, workflows, models, and many more.)

Read the details below, or join our Release Data Talks this Thursday on December 9 to chat directly to our developers about the newest features.

A core feature of KNIME is that it provides choice, so all data workers can use the tools they know and love best. In the last cycle, a lot of development work has gone into ensuring that Python is just as performant inside of KNIME as it is anywhere else.

With this release, see performance improvements of up to 50x, compared to previous versions.

This is the next step for KNIME in a series of developments focused on strengthening the relationship between a no-code and a scripted approach to data science.

Before any insight is generated, teams need to access and prepare data that is frequently spread across multiple databases and in different formats. Without a tool like KNIME, data workers spend the majority of their time here. To this end, KNIME is consistently adding and improving connectors, so you can access any data type from any data source.

This time around, you’ll find:

• Handy improvements to Excel nodes, which enable you to make changes to spreadsheets without affecting formatting.
• The new Generic S3 Connector which enables you to reach out to every kind of S3 compatible file system, for example MinIO.
• Improvements to the MongoDB extension, including a performance boost of up to 50 times than the single row execution used previously.

Frequently the work that a data worker does in a single workflow is made available to team members via an API. Today, we’ve released KNIME-native API endpoints, which allow workflows to much more efficiently be called by other workflows.

As a result:

• More data types can be shared, beyond what was possible with JSON (e.g., text documents, images, workflows, models, and many more.)
• You can now define multiple input & output nodes (of consistent or varying data types).

All features have been built based on community user feedback. Please let us know what you think, either by posting in the forum or by telling us during our Release Data Talks this Thursday.

See all the new features of this release on the What’s New page.