KDB.AI Versioning and Upgrading

This section explains how KDB.AI versioning works and provides information on how to interpret version numbers.

KDB.AI version

To learn more about the features included in each version, go to our latest or previous release notes.

Semantic versioning

KDB.AI adheres to semantic versioning (SemVer). This means that a version number follows the format MAJOR.MINOR.PATCH, where:

  • MAJOR version changes in case of significant backward-incompatible API changes, new major features or architectural overhauls. Users should pay attention to major version updates and test their existing workflows.

  • MINOR version increments for backward-compatible functionality additions and improvements to existing features. Users can safely upgrade to a new minor version without major disruptions.

  • PATCH version increases for backward-compatible bug fixes, security updates, and performance optimizations. Users are encouraged to keep their installations up to date with the latest patch versions.

Release cadence

  • Major releases: potentially once per year, when we introduce backward-incompatible changes.

  • Minor releases: 1 per quarter or 1 per month.

  • Patch releases: As needed, approximate historical cadence is 1 patch every 2-4 weeks.

Documentation versioning

The documentation is mapped against the KDB.AI Product releases. Let's assume you're looking for information about hybrid search:

  1. Navigate to the previous release notes.

  2. Find the release version when hybrid search was launched. You'll notice that we introduced it in the KDB.AI Server v1.1.0 release.

  3. Go to the top left dropdown on the documentation site and select 1.4.0 or an earlier version.

Release notes

For each release, KDB.AI provides detailed release notes that highlight the changes made in that version. These notes include:

  • A summary of new features and improvements.

  • Any known issues or limitations.

  • Upgrade instructions (if applicable).

Upgrade to latest version

To upgrade KDB.AI Server, run the following commands:

Shell

Copy 
export USERNAME=<Signup email address>
export BEARER=<Copied from your Welcome email>
docker login portal.dl.kx.com -u $USERNAME -p $BEARER
docker pull portal.dl.kx.com/kdbai-db:latest
pip install kdbai-client --upgrade

Troubleshooting

  • Make sure you already have your license and bearer token.

  • If you encounter connection issues, verify your network settings and access credentials.

If you need help, reach out to the KDB.AI Slack community or email support@kdb.ai.

Next steps