Observability has a data problem. AI is about to make it worse.
This title could be clearer and more informative.Try out Clickbait Shieldfor free (5 uses left this month).
Bronto, a Dublin-based observability startup, argues that the industry's real problem is telemetry storage economics rather than dashboards. Co-founders Trevor Parsons and Noel Ruane say OpenTelemetry solved instrumentation but exposed how expensive and compromised storage, retention, and search remain, forcing teams to sample data, shorten retention, or rehydrate cold data. They claim their custom polymorphic data store, BrontoDB, lets enterprises retain over 100x more full-fidelity data without slowdowns, solves high-cardinality metrics costs and log indexing trade-offs, and avoids trace sampling entirely, while billing for actual query usage rather than stored volume. They argue AI will worsen the data problem by generating more non-deterministic, trace-heavy telemetry that needs longer retention, making full data access essential for AI-driven operations.
