Edit

Materialized views in Azure Managed Instance for Apache Cassandra

  • Article

Azure Managed Instance for Apache Cassandra is a fully managed service for pure open-source Apache Cassandra clusters. The service also allows configurations to be overridden, depending on the specific needs of each workload, allowing maximum flexibility and control where needed. This article discusses how to enable materialized views.

Materialized view support

Materialized views are disabled by default, but users can enable them on their cluster. However, we discourage users of Azure Managed Instance for Apache Cassandra from using materialized views. They are experimental (see Materialized Views marked experimental-Apache Mail Archives and the proposal to do so). In particular:

  • The implementation of materialized views is distributed system design that hasn’t been extensively modeled and simulated. There have been no formal proofs about its properties.
  • There is no way to determine if a materialized view is out of sync with its base table.
  • There is no upper bound on how long it takes for a materialized view to be synced when there is a change to its base table.
  • If there is an error and a materialized view goes out of sync, the only way to repair it is to drop the materialized view and recreate it.

Microsoft cannot offer any SLA or support on issues with materialized views.

Alternatives to materialized views

Like most NoSQL stores, Apache Cassandra is not designed to have a normalized data model. If you need to update data in more than one place, your program should send all the necessary statements as part of a BATCH. This has two advantages over materialized views:

  • BATCH guarantees that all statements in the batch are committed or none.
  • All the statements have the same quorum and commit semantics.

If your workload truly needs a normalized data model, consider a scalable relational store like Azure's Hyperscale PostgreSQL.

How to enable materialized views

You need to set enable_materialized_views: true in the rawUserConfig field of your Cassandra data center. To do so, use the following Azure CLI command to update each data center in your cluster:

FRAGMENT="enable_materialized_views: true"
ENCODED_FRAGMENT=$(echo "$FRAGMENT" | base64 -w 0)
# or
# ENCODED_FRAGMENT="ZW5hYmxlX21hdGVyaWFsaXplZF92aWV3czogdHJ1ZQo="
resourceGroupName='MyResourceGroup'
clusterName='cassandra-hybrid-cluster'
dataCenterName='dc1'
az managed-cassandra datacenter update \
    --resource-group $resourceGroupName \
	--cluster-name $clusterName \
	--data-center-name $dataCenterName \
	--base64-encoded-cassandra-yaml-fragment $ENCODED_FRAGMENT

Next steps