Skip to content

Deploying

Deploying the kdb Insights Database microservice can be done via Docker or Kubernetes.

The KX SDK Reference Architectures contains docker and helm based reference architecture and deployment instructions. The aim is to allow users to quickly deploy kdb Insights SDK architectures to ingest and persist data to the kdb Insights Database as well as scale horizontally and deploy pipelines with the full feature set of the kdb Insights SDK toolset.

Docker

Docker deployments are typically orchestrated with Docker Compose:

  • Ingest and persist

    • Publish, persist and query data
    • Includes the necessary sidecars to provide metric data for the running services.

  • Ingest, transform and persist

    • Publish, transform, persist and query data
    • Builds on the previous architecture highlighting a typical deployment of the kdb Insights SDK components with the Stream Processor as a means to read in and transform data.

These architectures use the q language interface to publish data and REST or kdb+ to query the data. The Samples demonstrate how to publish and query using different interfaces.

Kubernetes (Helm)

Deploying the microservices can be done by using helm:

  • Ingest and persist

    • Publish, persist and query data
    • Includes the necessary sidecars to provide metric data for the running services.

  • Ingest, transform and persist

    • Builds on the previous architecture highlighting a typical deployment of the kdb Insights SDK components with the Stream Processor as a means to read in and transform data.

  • Sharded database

    • Deploys a multi-database, multi-shard application, providing a templated example of how to build a horizontally scalable database with kdb Insights SDK Services. It utilizes data sharding to ensure ingestion and persistence can scale as the amount of data being ingested expands in line with business sectors and use cases.