Home icon

Scaling cluster manager and admin APIs in Amazon OpenSearch Service

Big Data Blog



The article discusses performance improvements for cluster management and admin APIs in Amazon OpenSearch Service, specifically for scaling large clusters up to 1,000 nodes with 500,000 shards.

  • Identified and resolved four major bottlenecks in cluster state communication and admin API processing
  • Implemented a lightweight transport API to check cluster state freshness, reducing leader node CPU usage by 50%
  • Introduced local aggregation and filtering of statistics to reduce network throughput and memory pressure
  • Added transport layer cancellation for long-running admin API requests
  • Created paginated list APIs for metadata retrieval to handle large cluster sizes

These optimizations improve performance and stability for OpenSearch Service clusters of all sizes, ensuring efficient metadata management and reduced load on cluster manager nodes.



Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

Jul 3
2025
Enhance stability with dedicated cluster manager nodes using Amazon OpenSearch Service
Nov 21
2025
Introducing Cluster insights: Unified monitoring dashboard for Amazon OpenSearch Service clusters
Nov 19
2024
Amazon OpenSearch Service now scales to 1000 data nodes on a single cluster
Jan 7
2025
Use CI/CD best practices to automate Amazon OpenSearch Service cluster management operations

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.