Elasticsearch on Kubernetes - Auto scale

Hi,

We are upgrading our Elasticsearch clusters to work on top of EKS and we want to use autoscaling.

We wonder if/when you'll support the autoscaling based on memory or CPU, and if there's a plan to dynamically add/remove replicas for overloaded shards/indices (in term of number of requests).

Thanks,
Itay

1 Like

Hi,

ECK can already leverage the Elasticsearch autoscaling feature to adjust automatically the number of Pods and the allocated resources (see the documentation Elasticsearch autoscaling | Elastic Cloud on Kubernetes [2.5] | Elastic).

I'm not aware of a plan to dynamically add/remove replicas for overloaded shards/indices. Elastic Stack - Discuss the Elastic Stack would be the right place to ask.