Setting up data stores

Edit online

The self-hosted Instana backend requires multiple data stores for metrics, traces, logs, and other data. You can install data stores in your cluster by using third-party operators or by using Docker-based data stores on a single host.

Note: The use of Docker-based data stores for the Instana backend Operator is deprecated. However, if you are already using the Docker-based data stores, you can continue to use it.

To set up data stores as a high-performing and distributed data store cluster, use third-party Kubernetes operators.

IBM provides and maintains supported images of the following data stores for use with Instana Custom Edition:

  • Cassandra: 4
  • ClickHouse: 25
  • ClickHouseKeeper: 25
  • Elasticsearch: 8
  • Kafka: 4
  • Postgres: 15
  • Zookeeper: 3

For Linux on IBM Z and LinuxONE data store versions, see Installing data stores on Linux on IBM Z and LinuxONE.

For Linux on Power data store versions, see Installing data stores on Linux on Power (ppc64le).

For more information about installing the data store operators, see Installing third-party data store operators.

Note: For Self-Hosted Custom Edition (Kubernetes or Red Hat OpenShift Container Platform) with large metric loads, you must also deploy BeeInstana by using the BeeInstana Operator. The BeeInstana Operator creates, configures, and manages BeeInstana clusters that run on Kubernetes. For more information, see Using BeeInstana Kubernetes Operator.

Options for setting up data stores, as described in the following text.