DOCSEARCHbyAlgolia
Submit Docs

KNIME Documentation

website·Indexed 39 days ago

https://docs.knime.com/ap/latest

Comprehensive guides for the KNIME ecosystem, including the Analytics Platform for workflow development, Business Hub for enterprise deployment and governance, Community Hub for collaboration, and KNIME Server administration. Covers installation on Kubernetes and OpenShift, workflow validation, secret management, and database extensions.

Tokens
841K
Snippets
1.8K
Records
6.3K
Agent score
60%

What's inside KNIME

  1. Overview of KNIME Expression nodes

    KNIME provides three primary nodes that utilize the KNIME Expression Language for data manipulation. These nodes feature an expression editor with an input panel (left), a function catalog (right), and a control bar to sequence multiple expressions. Results can be previewed using the evaluation buttons.

  2. Overview of KNIME Business Hub deployment types

    The KNIME Business Hub supports four primary deployment options for workflows:

    • Data App: Deploy a workflow as an interactive application.
    • Schedule: Set workflows to execute automatically at specified times.
    • Service: Create an API Service to allow external calls to trigger the workflow.
    • Trigger: Implement deployments based on specific events.

  3. Overview of KNIME Extension types

    KNIME functionality can be expanded using extensions provided by different sources:

    • KNIME Extensions: Open source extensions developed and maintained by KNIME. This includes KNIME Labs Extensions, which provide cutting-edge but potentially non-finalized functionality.
    • Community Extensions: Industry or domain-specific functionality. These are split into Trusted (tested for backward compatibility and quality standards) and Experimental (direct from community developers).
    • Partner Extensions: Capabilities maintained by partners, which may be free or require a commercial license.

  4. Overview of the Pixi Backend for Python in KNIME

    The modern Python integration in KNIME is powered by Pixi, a package management tool built on the Conda ecosystem. It is encapsulated within KNIME, requiring no manual installation or user configuration. Key advantages include cross-platform environment resolution (Windows, Linux, macOS), faster node configuration, and the ability to define environments directly within the KNIME workflow.

  5. Overview of Vector Stores in KNIME

    A vector store maps documents to numerical vectors, enabling similarity-based search which is the foundation of retrieval-augmented generation (RAG). Vector stores are used to reuse embeddings across workflows, perform semantic search over documents, retrieve relevant context for RAG, and persist embeddings for scheduled use. If only immediate, in-memory processing is needed, use the 'Create embeddings' functionality instead.

  6. Overview of KNIME Community Hub

    The KNIME Community Hub (https://hub.knime.com) is a central knowledge base for the KNIME Community. It serves as a repository to find, upload, and organize nodes, extensions, components, and workflows. Users can use these assets as blueprints in the KNIME Analytics Platform to build data science workflows and collaborate with other users through projects.

  7. Overview of KNIME Business Hub Admin Workflows

    KNIME Business Hub provides a set of specialized data apps and workflows for global and team admins to manage infrastructure, users, and resources. Key capabilities include:

    • Monitoring: Track user/team activity, resource utilization (execution contexts, scheduling), content/storage management, performance/reliability, Gen AI usage, and deployment performance.
    • Diagnostics: List installed extensions, nodes, conda environments, and Linux packages; test URL accessibility and network/proxy issues from execution contexts; and inspect active jobs.
    • Maintenance: Sync deployment settings with execution context defaults, delete unused workflow versions or outdated executions, remove unhealthy deployments, and bulk-update deployments to the latest workflow version.

  8. Overview of AWS Infrastructure for KNIME Business Hub

    KNIME Business Hub supports offloading storage to AWS. Currently, AWS S3 and RDS are the only supported external/offloaded storage solutions. Note that external storage is only supported for new installations; automatic migration from existing deployments to external AWS storage is not supported. KNIME is not responsible for backup/restore operations of external Postgres and S3; users should utilize AWS native data protection services.

  10. Overview of Workflow Validation in KNIME Business Hub

    Workflow validation allows administrators to govern workflow quality by ensuring only compliant and reviewed workflows are used in production. This is achieved by assigning version labels (e.g., 'Approved', 'Not approved') to workflow versions and restricting execution contexts so that only versions with specific labels can be executed. This framework supports governance, quality assurance, and automated or human-in-the-loop review processes.

  11. Overview of Workflow Validation in KNIME Business Hub

    Workflow validation allows administrators to govern workflow quality by ensuring only compliant and reviewed workflows are used in production. This is achieved by assigning version labels (e.g., 'Approved', 'Not approved') to workflow versions and restricting execution contexts so that only workflows with specific labels can be executed. The system is modular, utilizing data apps, triggers, and components to enforce organizational standards, such as the 'four eyes' principle for peer review or automated compliance checks.

  12. Overview of KNIME Extension types

    KNIME functionality can be expanded using extensions provided by different sources:

    • KNIME Extensions: Open source extensions developed and maintained by KNIME. This includes KNIME Labs Extensions, which provide cutting-edge, potentially non-finalized functionality.
    • Community Extensions: Industry-specific functionality (e.g., bioinformatics, image processing). These are split into Trusted (tested for backward compatibility and quality standards) and Experimental (direct from community developers).
    • Partner Extensions: Capabilities maintained by partners, ranging from industry-specific tools to scientific software integrations. Some may require a commercial license.