Predictive Insights for IT Operations #bbuzz
summary
Describes a system that applies machine learning to datacenter event streams to surface correlated anomaly detection across thousands of time series, using heuristics (reduced event counts), finance-style signals like Bollinger bands, and more complex methods such as probabilistic graphical models. Integrates with Hadoop/HBase, Solr/Elasticsearch and Cassandra, visualizes normalized percentage changes and directed flow graphs, and shows it can filter billions of data points a day to highlight likely correlations for investigation.
AI-written summary · about
chapters (7)
- 0:00Introduction and background
- 3:42Complexity in modern data centers
- 5:07Learning from high-frequency trading
- 7:15Ad targeting and real-time systems
- 8:40Personalized recommendation systems
- 10:06Visualizing large-scale data
- 12:54Advanced visualization techniques
related talks
52mDevil in the Haystack
37mBig Data Analytics, Machine Learning, Real Time Processing
19mBig Data
43mBeyond Logs: Real-Time Anomaly Detection with Machine Learning
11mTransforming Problem-Solving with Data-Driven Insights
23mAI-Driven Enterprise Intelligence
42mMachine Learning In Observability Systems: From Insights To Action
32mBusiness Intelligence in Microservice architecture #bbuzz