Link Search Menu Expand Document Documentation Menu
Documentation

Optimizing benchmarks

You can optimize OpenSearch Benchmark results by distributing load across multiple machines, expanding data corpora, setting target throughput, and randomizing queries to simulate real-world traffic.

Running distributed loads

Configure multiple load generator machines to run large-scale benchmarks

Expanding a workload's data corpus

Increase dataset size and complexity for more realistic benchmark scenarios

Target throughput

Set specific throughput targets to match your production workload patterns

Performance testing best practices

Follow these best practices to achieve more meaningful benchmark results

Randomizing queries

Generate varied search patterns to simulate real-world queries

350 characters left

Have a question? .

Want to contribute? or .