Large-scale logging made easy
summary
VictoriaLogs is a single binary log database that accepts Filebeat/Fluentd/Logstash/Vector/Promtail input, exposes an HTTP API and a LogsQL query language, and integrates with jq/awk/cut for shell pipelines. It claims handles 100 TB on a single node with ~6 TB disk by using 30x compression vs Elasticsearch/Loki, replacing large clusters and returning JSON lines for billions of matches. Missing today: cluster mode, built-in advanced extraction/stats, and tighter Grafana integration; the single-node system otherwise works as described.
AI-written summary · about
chapters (13)
- 0:00Introduction to logging
- 1:16Purposes of logging
- 4:01Traditional logging explained
- 7:16Disadvantages of traditional logging
- 8:40Large-scale logging principles
- 9:24Cloud vs on-prem solutions
- 11:31Comparing on-prem solutions
- 17:57VictoriaLogs features
- 20:22Command line integration demo
- 29:35Why VictoriaLogs excels
- 31:23VictoriaLogs drawbacks
- 32:46Recap and summary
- 33:55Q&A session
related talks
40mLarge-Scale Logging Made Easy
01h 00mApplication Logging & Logstash
30mHow We Scaled Metrics Monitoring for 10,000 Applications With Victoria...
46mBuilding a scalable logging platform
39mFast log management for your infrastructure
14mCentralized Logs using ELK (Elasticsearch, Logstash, Kibana) Stack
43mPractical Application Logging and Monitoring
28mFrom Fragmented Logs to Unified Insights: Scaling O11y