IoTDB - Moving from Relational to Timeseries
summary
A remote-monitoring system migrates from H2/Postgres prototype to IoTDB because 10 hosts produce 5 GB in 30 days and queries take minutes; after migration 50 hosts and 90 days use 1.24 GB. The stack is Java agents sending 15s CPU/RAM/disk/network samples via JDBC (moving toward the Java API and TSFile writes), a custom web front end, and use of auto create schema and align by device plus down-sampling. It runs fast, but has no UPDATE, needs sequential inserts, and required schema and path naming changes.
AI-written summary · about
chapters (11)
- 0:00Introduction and project
- 1:14Moving to IoTDB
- 4:16Features of IoTDB
- 7:59Anatomy and schema design
- 13:38Getting started with IoTDB
- 17:56Data ingestion methods
- 21:43Querying basics
- 27:56Advanced querying techniques
- 33:44Analysis and visualization
- 38:30Lessons and performance
- 42:41Conclusion and Q&A
related talks
49mData management and Analytics for Industrial IOT
52mDistributed IoT Telemetry Storage Based on PostgreSQL and TimescaleDB
10mBuilding IoT Apps with JSONB & Timeseries
49mThinking in time series, the new solution to your pro...
44mOpenTSDB - Time Series Schema on Schemaless NoSQL
07mTime Series Databases
58mIntroduction to Timeseries Databases
14mBuilding the future of time series data