Preparing an index cluster deployment
You need to make some preparations before you can deploy the Qlik Big Data Index cluster with Helm.
Configuring the context
You need to ensure that your context is set up with kubectl. If needed, you can list your contexts and change between contexts.
Checking which context is active
Listing your contexts
Changing active context
Setting up the Helm chart repository
You need to make sure that you can access the repository where the Helm charts containing the Qlik Big Data Index applications are stored.
-
Add the Qlik Big Data Index chart repository locally with Helm:
$ helm repo add qlik https://qlik.bintray.com/qabdicharts -
Update the repository
$ helm repo update -
Check that you can see the Qlik Big Data Index Helm chart.
$ helm search qlik/bdi
You should now see the Qlik Big Data Indexchart.
Storage provision
There are some considerations you need to make when selecting which file system to use and how it is provisioned.
AWS
When you use Amazon EFS, it is possible to utilize burst credits setup. If the EFS instance runs out of burst credits , performance for Qlik Big Data Index is drastically reduced if both index and data are stored on EFS.
https://aws.amazon.com/premiumsupport/knowledge-center/efs-burst-credits/
Static storage provision
It is possible to provision volumes statically or dynamically. When you provision storage statically, make sure that the disk is big enough for your intended data amounts.
Non-managed deployment
We do not recommend using CIFS protocol based storage servers, for example Microsoft Windows Share folders. Qlik Big Data Index cannot store symbol server outputs in CIFS protocol based storage servers.