Release Notes: Time Series Analytics 2025#

Version 2025.2#

December 10, 2026

This release introduces comprehensive configuration improvements, GPU acceleration support, and enhanced security measures for the microservice. It offers two deployment options:

  • Docker compose deployment on single node

  • Helm deployment on kubernetes single cluster node

New

  • GPU Acceleration: Added GPU device support with Intel oneAPI integration for improved inference performance

  • Enhanced Documentation: New comprehensive configuration guides for UDFs, MQTT alerts, and OPC UA alerts

  • Device selection support for CPU or GPU inference

Improved

  • Docker Optimization: Upgraded to Kapacitor 1.8.2 with multistage builds for improved efficiency and security

  • New “How to Configure” guide with example JSON configurations

  • DockerHub documentation for Docker images and Helm charts

  • Upgraded base Docker image from Kapacitor 1.7.7 to 1.8.2

  • Enabled multistage Docker builds reducing image size

  • Added Nginx root URL routing support

  • Updated Helm charts and deployment configurations

  • Improved HTTP status code handling (400, 422, 503) for better error reporting

  • Standardized logging format using parameterized strings

  • Removed deprecated Model Registry references

  • Cleaned up documentation structure across components

  • Removed oneAPI toolkit to reduce image size

Fixed

  • Fixed Trivy security vulnerabilities by updating FastAPI and Kubernetes configurations

  • Resolved bandit security vulnerability for tmp directory usage

  • Fixed Python linting issues with comprehensive docstrings

  • Fixed OPC UA alert error code propagation

  • Corrected documentation links and architecture references

  • Updated OPC UA server certificate naming

  • Fixed variable naming and removed duplicate imports

Upgrade Notes

  • Docker images now use Kapacitor 1.8.2 - UDF implementations updated for API compatibility

  • Helm chart version updated from 1.0.0 to 1.1.0-weekly

  • Python dependencies updated across multiple licenses

More details at user-guide

Version v1.0.0#

August, 2026

This is the first version of the Time Series Analytics microservice. It offers two deployment options:

  • Docker compose deployment on single node

  • Helm deployment on kubernetes single cluster node

New

  • Bring your own Data Sets and corresponding User Defined Functions(UDFs) for custom analytics: Easily implement and deploy your own Python-based analytics logic, following Kapacitor’s UDF standards.

  • Seamless Integration: Automatically stores processed results back into InfluxDB for unified data management and visualization.

  • Model Registry Support: Dynamically fetch and deploy UDF scripts, machine learning models, and TICKscripts from the Model Registry microservice, enabling rapid customization and iteration.

  • Versatile Use Cases: Ideal for anomaly detection, alerting, and advanced time series analytics in industrial, IoT, and enterprise environments.

More details at user-guide

On this page

This Page