Run log analytics for a fraction of the cost with the new engine for Amazon OpenSearch Service

Amazon OpenSearch Service introduces a purpose-built log analytics engine that significantly reduces costs and improves performance for observability workloads at scale.

• Up to 4x better price-performance compared to the general-purpose engine while retaining full-text search capabilities
• Up to 2x faster analytical queries using vectorized execution on columnar data
• Up to 2x higher ingestion throughput with append-only columnar storage reaching 1.78M docs/sec
• Up to 70% lower storage costs; retain 3x more data at the same cost with 80% compression versus raw JSON
• Uses Apache Parquet format with Lucene inverted index for selective lookups and search
• Supports PPL and SQL query languages; DSL queries not supported at launch
• Created as domain-level setting at creation time; requires new domain migration
• Benchmarked at billion-document scale across OTEL traces, logs, and web server access logs
• Generally available in all commercial AWS Regions where OpenSearch Optimized Instances are available

The optimized engine enables organizations to retain more observability data within budget constraints while accelerating analytical queries through columnar storage and vectorized execution.