From Overwhelmed to Optimized - How We Cut Envoy Logging Costs by 65% While Boosting Throughput 60%

Learn how to dramatically reduce Envoy logging costs while improving system performance in this 14-minute conference talk from the Cloud Native Computing Foundation. Discover practical strategies for optimizing telemetry at production scale when traditional observability approaches consume excessive infrastructure budget and degrade system performance. Explore the implementation of intelligent filtering, sampling strategies, and selective metric collection techniques that maintain full operational visibility while significantly reducing overhead. Understand how these optimizations achieved a 65% reduction in logging infrastructure costs and 60% improvement in overall throughput while maintaining 99.9% uptime and reducing mean time to resolution from hours to minutes. Gain insights into open source contributions made to Envoy for improving observability and learn about future developments including the use of Access Logging Service for internal system integration at Databricks.