Loading

Reduce replicas on content data tier

Some indices on the content data tier carry more replica shards than your workload needs. Extra replicas use CPU, memory, and disk on every content node and can slow indexing.

Note

For a complete list of insights, refer to AutoOps insights.

Field Value
Component Elasticsearch
Severity Low
Scope Node
Domains replication, resource-utilization

The following is an example of what you might see when this insight is triggered. Real insights use live data and links from your deployment or cluster.

There was 1 replica detected in content data tier for the following indices: logs-prod-000045, logs-prod-000046.

Note

AutoOps shows different recommendations depending on how their conditions match your deployment or cluster.

Reducing the number of replicas for the content tier can improve performance, save on storage costs, enhance indexing throughput and ensure cluster stability. This is because replicas introduce additional resource usage, which can negatively impact performance and stability.