Deploy on a Kubernetes Cluster

This article demonstrates how to deploy a ByConity cluster on a Kubernetes (k8s) environment.

Environment Setup

• Install the Kubernetes command-line tool kubectl locally to manage the Kubernetes cluster.
• Install helm, the package manager for Kubernetes applications.
• Clone the bconity-deploy repository locally.

git clone git@github.com:ByConity/byconity-deploy.git
cd byconity-deploy

Deployment Steps

1. Acquire a Kubernetes Cluster

   You can either deploy or purchase a Kubernetes cluster. Refer to the official documentation for more details.

2. Prepare Storage

   For optimal TCO and performance, it's recommended to use local disks alongside ByConity server and worker nodes. Storage solutions like OpenEBS local PV can be utilized. Note: Storage for ByConity server and worker nodes is only used for disk caching and can be deleted at any time.

3. Configure your Helm File

   Copy and modify certain fields from ./chart/byconity/values.yaml such as:

   • Storage class name
   • Timezone
   • Number of server/worker replicas
   • HDFS storage configuration

   If you want to use your own HDFS cluster, set hdfs.enabled to false and override the HDFS configuration in the values.yaml file.

   byconity:
     hdfs_address: hdfs://your-own-hdfs:port
   hdfs:
     enabled: false

   For FDB (FoundationDB) configuration: If you want to use your own FDB, set both fdb.enabled and fdb-operator.enabled to false. You can refer to values_use_existing_fdb.yaml for configuration.

   byconity:
     hdfs_address: hdfs://byconity-hdfs-namenodes:8020 # or use your own hdfs
     use_existing_fdb: true
     fdb_cluster_file: your-fdb-cluster-file-content # e.g., byconity_fdb:Is0hBgl6iICdHuspBmhAODmD5WISXKzI@192.168.224.150:4501,192.168.226.83:4501,192.168.228.152:4501
   fdb:
     enabled: false
   fdb-operator:
     enabled: false

4. Initialize the ByConity Cluster

   # Install with fdb CRD first (if using custom FDB setup)
   helm upgrade --install --create-namespace --namespace byconity -f ./your/custom/values.yaml byconity ./chart/byconity --set fdb.enabled=false
   
   # Install with fdb cluster (if using ByConity's FDB setup)
   helm upgrade --install --create-namespace --namespace byconity -f ./your/custom/values.yaml byconity ./chart/byconity

   Check and wait for all Pods to be ready:

   kubectl -n byconity get pods

   Once the installation is successful, you can start using ByConity!

   $ kubectl -n byconity exec -it sts/byconity-server -- bash
   root@byconity-server-0:/# clickhouse-client
   
   172.16.1.1 :)

5. Deleting the ByConity Cluster

   To delete the ByConity cluster, run:

   helm uninstall --namespace byconity byconity

6. Testing

   You can run some simple queries to test the cluster:

   CREATE DATABASE IF NOT EXISTS test;
   USE test;
   DROP TABLE IF EXISTS test.lc;
   CREATE TABLE test.lc (b LowCardinality(String)) ENGINE = CnchMergeTree ORDER BY b;
   INSERT INTO test.lc SELECT '0123456789' FROM numbers(100000000);
   SELECT count(), b FROM test.lc GROUP BY b;
   DROP TABLE IF EXISTS test.lc;
   DROP DATABASE test;

Make sure to replace any placeholders (like ./your/custom/values.yaml) with the actual path to your customized Helm values file.