Skip to main content

ClickHouse on EKS Stack

ClickHouse deployment on Amazon EKS — a high-performance, column-oriented OLAP database for real-time analytics on petabyte-scale datasets. This stack provisions a sharded, replicated ClickHouse cluster managed by the ClickHouse Kubernetes operator with a dedicated ClickHouse Keeper ensemble for replication and distributed DDL coordination.

Getting Started

1

Deploy Infrastructure

Provision the EKS cluster, Karpenter node pools, and the ClickHouse operator with ArgoCD

2

Launch a ClickHouse Cluster

Deploy a sharded, replicated ClickHouse installation backed by ClickHouse Keeper

3

Load Sample Data

Ingest the ClickHouse hits dataset from S3 into a Distributed/ReplicatedMergeTree table

4

Query and Test Failover

Run analytical queries, inspect index usage with EXPLAIN, and validate replica failover

Sample Workload: Hits Dataset

Load the canonical ClickHouse hits Parquet dataset from S3 into a Distributed table over ReplicatedMergeTree, run analytical queries, and demonstrate replica failover by deleting a pod.

ExampleOLAPS3Replication