stable/zeppelin/README.md
As of Nov 13, 2020, charts in this repo will no longer be updated. For more information, see the Helm Charts Deprecation and Archive Notice, and Update.
Zeppelin is a web based notebook for interactive data analytics with Spark, SQL and Scala.
This chart is deprecated and no longer supported.
To install the chart:
$ helm install stable/zeppelin
The following table lists the configurable parameters of the Zeppelin chart and their default values.
| Parameter | Description | Default |
|---|---|---|
zeppelin.image | Zeppelin image | dylanmei/zeppelin:{VERSION} |
zeppelin.resources | Resource limits and requests | limits.memory=4096Mi, limits.cpu=2000m |
spark.driverMemory | Memory used by Spark driver (Java notation) | 1g |
spark.executorMemory | Memory used by Spark executors (Java notation) | 1g |
spark.numExecutors | Number of Spark executors | 2 |
hadoop.useConfigMap | Use external Hadoop configuration for Spark executors | false |
hadoop.configMapName | Name of the hadoop config map to use (must be in same namespace) | hadoop-config |
hadoop.configPath | Path in the Zeppelin image where the Hadoop config is mounted | /usr/hadoop-2.7.3/etc/hadoop |
ingress.enabled | Enable ingress | false |
ingress.annotations | Ingress annotations | {} |
ingress.hosts | Ingress Hostnames | ["zeppelin.local"] |
ingress.path | Path within the URL structure | / |
ingress.tls | Ingress TLS configuration | [] |
nodeSelecor | Node selector for the Zeppelin deployment | {} |
The Hadoop chart can be used to create a YARN cluster where Spark jobs are executed:
helm install -n hadoop stable/hadoop
helm install --set hadoop.useConfigMap=true,hadoop.configMapName=hadoop-hadoop stable/zeppelin
Note that you may also want to set the
spark.numExecutorsvalue to match the number of yarn NodeManager replicas and theexecutorMemoryvalue to half of the NodeManager memory limit.