
Honeycomb Product and Custom Database

#36 - Building High-Performing Teams with Observability and CI/CD - Charity Majors
Topic 3 of 6
Published Apr 26, 2021 · Clip 20:26–31:18 (10:52)
Charity explains how Honeycomb's debugging flow works: instead of jumping between metrics, logging, and tracing tools, users draw a box around an anomaly and BubbleUp diffs dimensions to surface the answer. She describes a surprising customer use case, the build-events library that instruments CI/CD pipelines as traces to visualize where build time goes. She then explains why Honeycomb wrote its own columnar database, prioritizing sub-second query latency and schemaless, index-free storage, and how the architecture evolved to become serverless with data aging out to S3 and Lambda-based queries.