New log analytics engine for Amazon OpenSearch Service cuts cost up to 70%

AWS introduced a purpose-built log analytics engine for Amazon OpenSearch Service, targeting one of the most expensive workloads in modern observability: high-volume log storage and analysis. The new engine delivers up to 4x price performance, 2x faster ingestion, up to 2x faster analytical queries, and up to 70% lower storage costs — all without sacrificing search capabilities on the same data.

Log analytics is a large and growing cost center as applications, and increasingly AI systems, generate massive telemetry volumes. Traditional search-optimized engines can be expensive for log-scale data because they index everything for fast full-text search even when most logs are queried analytically or rarely at all. A purpose-built engine can optimize storage layout and query paths for log-specific access patterns, driving the dramatic storage and cost reductions AWS cites.

The 70% storage-cost reduction is the headline number, directly addressing customer cost anxiety that surfaced repeatedly this week — from Uber burning its AI budget in months to a $55k Gemini API bill from key abuse. AWS is positioning cost efficiency as a core platform value, complementary to its compute-side moves like Graviton5 and Lambda MicroVMs.

The release is part of a broad AWS operational and cost-tooling push including ECS zone-aware routing (cutting cross-AZ transfer costs), CloudWatch alarms from log queries, and RDS cross-region backup expansion. Together they reflect a strategy of winning enterprise workloads on total cost of ownership, not just capability. Watch for independent benchmarks validating the 4x price-performance claim and how the new engine compares to dedicated log platforms like Datadog and Splunk on cost at scale.