Google Cloud BigQuery Dataset¶
Google Cloud BigQuery is a service that provides a relational database that is optimized for analytical workloads. It is a good choice for storing data that is relational in nature.
NAIS provides BigQuery for storing and working with analytical datasets as a managed service through Google Cloud Platform. This page describes how to get started with BigQuery for your applications.
NAIS Application yaml manifest options¶
Full documentation of all available options can be found over at: spec.gcp.bigQueryDatasets[]
.
Example of an application using a nais.yaml
provisioned BigQuery Dataset can be found here: testapp.
Caveats to be aware of¶
Once a BigQuery Dataset is provisioned, it will not be automatically deleted - unless one explicitly sets spec.gcp.bigQueryDatasets[].cascadingDelete
to true
.
Clean up is done by deleting application resource and deleting the BigQuery instance directly in console.cloud.google.com.
When there exist no tables in the specified BigQuery Dataset, deleting the "nais application" will delete the whole BigQuery Dataset, even if spec.gcp.bigQueryDatasets[].cascadingDelete
is set to false
.
The name of your Dataset must be unique within your team's GCP project.
The NAIS Manifest does not currently support updating any setting of existing BigQuery Datasets.
Thus, if you want a read-connection to a already-created BigQuery Dataset, take a look at nais/dp.
Since Kubernetes does not permit underscores (_
) in the names of any K8s resource, any underscores will be converted to hyphens (-
).
Example with all configuration options¶
See full example.