Move ElastiCache to Graviton
AWS Resource Type:
ElastiCache allows to run managed clusters of Redis and Memcached. Redis is an in-memory data store that works very well for use cases where ephemeral data needs to be accessed at low latencies, like pub/sub or stream processing workflows streams, and as a cache layer for higher latency persistent storage like relational databases. Migrating such ElastiCache clusters, so that they can use EC2 instances powered by Graviton2 processors would help in your AWS cost savings as it will eventually also lead to right-sizing of the clusters once migrated.
For each of the ElastiCache clusters in your environment, CloudFix checks:
- It's engine version
- Instance type
reserved-memory-percentfrom the cluster's parameter group.
This Graviton upgrade is an in-place upgrade. For the cases where the engine version is lower than 5.0.6, the first step is upgrading the engine version. There are minimal to no risks when performing this engine version update and hence, CloudFix has automated this process. After this, the upgrade of the instance type is performed.
For clusters running version 5.0.6 or later, only the step to upgrade the instance type is performed. If CloudFix finds that this doesn't require a maintenance window, then the update it applied immediately.
Potential savings (range in % on annual basis):
For regular EC2 instances, the cost difference between Gen 5 instances compared to a Gen 6 is 20%. Older generations are increasingly more expensive, so the potential savings from migrating to Graviton2 is bigger. For
cache* type instances, the price difference is around 5% savings.
However, once right-sizing is performed, we can expect improved savings, as Graviton2 can deliver higher performance even for equivalent instances, so once enough metrics are in place, we would be able to reduce instance size or count.
A Redis benchmark shows that a Graviton2 instance does provide a significant increase in requests per second for most tested operations (on average, 30% more).
Can CloudFix implement the fix automatically once I accept the recommendation?
Yes, for all the cases where the CloudFix Finder finds the opportunities, the ElastiCache clusters will be moved to Graviton2 automatically.
Is it possible to roll back once CloudFix implements the fixer?
No, rollback of this Fixer is not possible.
Does this fix require downtime?
For clustered Redis there is no downtime. For the non-clustered Redis, minimal downtime of a few seconds on the primary node is expected. For cases where there is a high traffic, CloudFix waits for a maintenance window to perform the updates.
Do it yourself (DIY) Instructions:
If you have a CloudFix subscription, CloudFix will execute these automatic cost savings for you. However, if you'd like to do it yourself - please follow the manual steps in the Migrate Amazon ElastiCache to Graviton2 processors article.
|Finding cost-saving recommendations with CloudFix is FREE! You'll need a paid subscription only if you'd like CloudFix to execute those recommendations.|