# 46% fewer nodes, 2x faster indexing: Confluent on Amazon OpenSearch Service

DevFeed: [46% fewer nodes, 2x faster indexing: Confluent on Amazon OpenSearch Service](<https://devfeed.tech/articles/46-fewer-nodes-2-faster-indexing-confluent-on-amazon-opensearch-service-127631.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/big-data/46-fewer-nodes-2x-faster-indexing-confluent-on-amazon-opensearch-service/>)

Author: Sukanth Rajan

Published: 2026-11-10T19:56:57Z

Content type: news

Language: en

Sources: [AWS Big Data Blog](<https://devfeed.tech/sources/aws-big-data-blog.md>)

Topics: [Amazon OpenSearch Service](<https://devfeed.tech/topics/amazon-opensearch-service.md>), [opensearch](<https://devfeed.tech/topics/opensearch.md>), [observability](<https://devfeed.tech/topics/observability.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [amazon-elastic-block-store-amazon-ebs](<https://devfeed.tech/tags/amazon-elastic-block-store-amazon-ebs.md>), [amazon-opensearch-service](<https://devfeed.tech/tags/amazon-opensearch-service.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [amazon-simple-storage-service-amazon-s3](<https://devfeed.tech/tags/amazon-simple-storage-service-amazon-s3.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [analytics-workloads](<https://devfeed.tech/tags/analytics-workloads.md>), [apache-kafka](<https://devfeed.tech/tags/apache-kafka.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [backpressure](<https://devfeed.tech/tags/backpressure.md>), [confluent](<https://devfeed.tech/tags/confluent.md>), [cost](<https://devfeed.tech/tags/cost.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [fleet](<https://devfeed.tech/tags/fleet.md>), [indexing](<https://devfeed.tech/tags/indexing.md>), [latency](<https://devfeed.tech/tags/latency.md>), [logs](<https://devfeed.tech/tags/logs.md>), [node](<https://devfeed.tech/tags/node.md>), [nodes](<https://devfeed.tech/tags/nodes.md>), [observability](<https://devfeed.tech/tags/observability.md>), [opensearch](<https://devfeed.tech/tags/opensearch.md>), [write](<https://devfeed.tech/tags/write.md>)

## AI overview

Confluent moved its Amazon OpenSearch Service logging hot tier from r6gd to oi2 using a two-phase, measure-first migration. At the same node count and ingest, oi2 used roughly one-third the CPU; iterative rightsizing then reduced nodes by about 46%, with each remaining node carrying about 1.9x the load. Indexing latency roughly halved and write queues fell below about one queued operation across four domains. JVM heap, not CPU, limited further reductions; one heavily downsized domain retained higher CPU and the only queue not near zero, so it needs monitoring through peak traffic. The final state had no write rejections and healthy clusters. Despite a higher per-node price, fewer nodes yielded an estimated 18% lower hot-tier cost at on-demand rates; a 1-year reserved footprint is estimated at 47-51% below the original on-demand baseline. These approximate figures exclude unchanged storage and dedicated-master tiers.

## Source excerpt

How Confluent modernized its Amazon OpenSearch Service observability hot tier, ingesting hundreds of terabytes of logs per day, by migrating from r6gd to the purpose-built oi2 instance family. A measure-first approach cut the fleet 46%, roughly halved indexing latency, and cleared write backpressure, lowering cost even at a higher per-node price.