ElastiCache Reduce Read Replicas

Here is the support article for the new ElastiCacheReduceReadReplicas fixer:

ElastiCacheReduceReadReplicas

Opportunity Name:
ElastiCacheReduceReadReplicas

AWS Resource Type:
Amazon ElastiCache (Redis and Valkey, non-clustered mode)

Opportunity Description

This CloudFix Finder and Fixer pair identifies and safely removes surplus read replicas from Amazon ElastiCache clusters using Redis or Valkey engines. These replicas often accumulate beyond actual performance needs, driving unnecessary costs. To ensure safety, the Fixer automatically creates and verifies a backup before performing any reduction.

Memcached clusters are explicitly excluded because they lack snapshot capabilities, making automated changes unsafe. The fixer only supports Redis and Valkey clusters in non-clustered mode.

Criteria for Identifying the Opportunity

The opportunity is considered when:

• The cluster uses the Redis or Valkey engine (case-insensitive).

• The cluster is non-clustered (ClusterEnabled = false).

• The cluster is in an available state.

• No pending modifications are present.

• The number of read replicas exceeds the target minimum (typically more than one).

Clusters running Memcached or with Cluster Mode Enabled are excluded.

Potential Savings (if known)

Savings depend on the number and size of removed read replicas. Each unnecessary read replica removed can significantly reduce monthly ElastiCache costs, especially in production environments.

What Happens When the Fixer Is Executed?

1. Snapshot Creation:
   A manual snapshot of the replication group is taken to allow recovery in case of failure.

2. Snapshot Validation:
   The fixer waits for the snapshot to be fully created and marked as "available".

3. Replica Reduction:

   • If the new replica count is zero, and Multi-AZ is enabled, it is first disabled.

   • The fixer then uses the DecreaseReplicaCount API to reduce replicas to the target number.

4. Post-Modification Validation:
   Verifies that the cluster returns to a healthy state and confirms the expected number of replicas.

Is It Possible to Roll Back Once CloudFix Implements the Fixer?

Yes. The backup created before modification enables manual restoration if necessary.

Can CloudFix Implement the Fix Automatically Once I Accept the Recommendation?

Yes. This fixer is fully automated. Once approved, CloudFix handles all validation, backup, and replica modification steps.

Does This Fix Require Downtime?

No. The replica reduction process is handled with AWS best practices and typically does not disrupt service.

Additional Resources

• What is Amazon ElastiCache? (AWS)
• Amazon ElastiCache API Reference
• Best Practices for using Read Replicas