Thinking in time series, the new solution to your pro...
summary
Advocates using time-series databases (InfluxDB, Timescale, Prometheus, Warp 10, VictoriaMetrics and Cassandra-based systems) as the right model for monitoring, fraud detection, signal analysis and large IoT fleets, and presents real deployments: a ventilator ingesting setpoints and sensor streams to detect breathing anomalies, an A350 flight test with ~700,000 parameters, and a cluster serving nearly a billion series at ~1M ops/sec. Compares database trade-offs, describes bucketization and sampling strategies, explains resampling/interpolation, warns about cardinality explosion and expensive compaction on deletion, and announces a new Tardis storage service.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 2:25What is a time series
- 6:55Discrete systems and continuity
- 10:07Sensor issues
- 16:26Time-series databases
- 21:41Time-series modeling
- 27:45Medical monitoring example
- 32:40Synchronization and interpolation
- 35:54Advanced time-series analysis
- 40:09Time-series use cases
- 42:15Data management and compliance
- 44:01Tardis service
- 45:01Q&A
related talks
58mIntroduction to Timeseries Databases
23mTime series database: Should I use one?
39mA New Data Model for Time Series Data Processing
07mTime Series Databases- 39mA New Data Model for Time Series Data Processing
26mTime Series Databases and Web Based Clients: Essential Tools...
47mBack to the Future of Time Series Data
30mCatching Waves with Time-Series Data