Rocketgraph shrinks huge logs so an LLM can debug them
Rocketgraph turns large amounts of service logs into a much smaller set of repeated patterns. The goal is to reduce the need to manually inspect Grafana or write LogQL searches when something breaks. problems often come from a schema mismatch, a database connection issue, or one unusual log line hidden inside millions of normal lines.
Rocketgraph gives logs a , groups similar lines together, and uses to score which patterns look unusual. It looks at signals such as how often a pattern appears, how similar the text is, and other feature values. In a typical case, it aims to shrink one million logs into about 200 to 300 patterns.
An LLM can then inspect that smaller snapshot instead of receiving the full stream of raw logs.
Key points
- Rocketgraph groups many logs into a smaller set of patterns.
- It aims to reduce one million logs to about 200 to 300 patterns.
- It gives each pattern an anomaly score so unusual behavior stands out.
- It is meant to reduce manual work in Grafana and LogQL.
- It focuses on hidden issues such as schema mismatches, database connection failures, and new log lines.