Big Spatiotemporal Data Management Using MobilityDB & Citus
summary
MobilityDB + Citus on Azure stores and queries continuous trajectories (tgeompoint) and achieves dramatic compression and scale: converting flat logs to temporal types gives 450 percent lossless compression, a demo cluster holds 44 million points / 35,000 trajectories, and Citus shards trips by vehicle id while small spatial tables are copied to workers. Demo queries show temporal, spatial, spatiotemporal range, kNN and that distributed spatiotemporal joins currently require reshuffling or ongoing Citus work.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:15What is MobilityDB
- 12:49Mobility DB Demo
- 14:08Temporal range query
- 16:43Spatial range query
- 17:45Spatiotemporal range query
- 19:29Spatial kNN query
- 20:19Spatiotemporal join query
- 21:37Spatiotemporal Data Partitioning Challenges
related talks
24mCitus from the Customer eyes
55mBreakout Session
31mCitus and JSON for real-time analytics at Vizor Games
16mOnline Scale-Out and Cloud Data Ingestion
16mKeynote – Umur Cubukcu
15mAn Efficient Cloud Storage Model with Compacted Metadata Management for Performance...
03h 30mAmericas Live Stream
25mSelf Hosted Citus using Cloud Computing