Large-Scale Logging Made Easy
summary
Claims an open source system named Victoria Logs that runs on a laptop and in production, stores terabytes by using 50× compression on typical Kubernetes logs, and needs 30× less memory and ~15× less disk than Elasticsearch. Shows cost math: Datadog ~$260,000/month and Grafana Cloud ~$50,000/month for 100 TB retained one month, while a laptop costs ~$100/month. Demonstrates Bloom filters, time partitions, column storage, and claims needle‑in‑haystack queries are ~1,000× faster than Loki.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 2:17Logging purpose and types
- 6:57Large-scale logging
- 10:41Big log amounts
- 17:18Log storage costs
- 17:39Why specialized logging databases are efficient
- 23:25Real-world logging examples
- 26:40Need large-scale logging?
- 27:47Victoria Logs
- 30:58Q&A
related talks
35mHow to Efficiently Manage Logs in Large-Scale Kubernetes Clusters
38mLarge-scale logging made easy
46mConquering Distributed Cloud Logging Without Breaking the Bank
28mFrom Fragmented Logs to Unified Insights: Scaling O11y
31mMy Logs Cost More Than My Services
53mGetting started with Grafana Loki, a modern logs database
01h 02mCentralised Logging without the blood, sweat and tears
41mScaling Observability: From Chaos to Clarity with Open Source Tools