Swiss Transport in Real Time: Tribulations in the Big Data Stack
summary
Proof-of-concept builds a near‑real‑time Swiss public-transport pipeline that ingests SBB APIs, pushes events via Kafka and Logstash, transforms state with Akka actors per train and Play/Scala, stores months of events in Elasticsearch storage, and renders interactive maps with React + D3 + Pixi. Demo handles thousands of vehicles (fast‑forwarded 2,000), shows station delay percentages and crowding forecasts, uses Jupyter/Python for a posteriori analysis, runs from Docker on GCE/AWS, and reports ~50× speedup vs raw HTTP with low CPU use — workable but not production‑hardened.
AI-written summary · about
chapters (6)
- 0:00Start
- 6:22Demo of the Real-Time Application
- 7:12Agenda
- 9:58Dispatching Events
- 11:10Open Data Transport Api
- 32:54Demo
related talks
42mPrinciples and Patterns for Streaming Data Analysis
30mIntroduction to Data Streaming
53mIntroduction to data streaming
46mSmart Transit: Real-Time Transit Information with FLaNK
38mOn Track with Apache Kafka
31mBuilding Real-time Pipelines with FLaNK: A Case Study with Transit Data
36mNext generation plug & play IT-systems for public transport
16mTidy Transit: Real Life Data Modeling for Public Transportation