Finding Weather & Traffic Patterns by Harness the Power of GPU Analytics

A rapid increase in population in the San Francisco Bay area has placed a strain on traffic flow on the major freeways which connect the region. In fact, San Francisco recently ranked 5th in the world for worst traffic congestion, leading to drivers spending countless useless hours on the road. Along with heavy traffic, adverse weather can further negatively impact driving conditions and cause heavy delays. Understanding traffic flow in different traffic situations and weather scenarios is the first step in reducing San Francisco’s traffic issue. A deep understanding of traffic could also benefit commuters as well as emergency services needing to reach their destination as quickly as possible.
However, analyzing large amounts of spatiotemporal data such as traffic and weather data has usually been considered tedious, difficult, and time-consuming. In this presentation, we’ll demonstrate how to use software tools from OmniSci, the pioneer in GPU-Accelerated analytics, to perform in-depth analysis on San Francisco’s traffic flow and find significant correlations between traffic congestion and weather.
By using the Immerse visualization client, we’re able to explore how traffic is significantly affected by severe weather at peak travel times. OmniSci’s Core database makes uncovering spatiotemporal insights of millions of data points in real-time effortless, which allows us to identify key takeaways of San Francisco traffic. During the talk, we’ll interactively identify which freeways to avoid in heavy rain scenarios and what other weather phenomena could lead to traffic congestion. Finally, we’ll show how data processing and preparation can be done iteratively with visualization by connecting python with OmniSci’s database.

Transcript

1. Finding Weather & Traffic Patterns by Harness the Power of GPU Analytics Budapest Data Forum | May 22, 2019
2. © OmniSci 2018 Aaron Williams VP, Global Community @_arw_ aaron@omnisci.com /in/aaronwilliams/ /williamsaaron Abraham Duplaa Developer Advocate Consultant abraham.duplaa@cdtm.de /in//abraham-duplaa/ slides: https://speakerdeck.com/omnisci
3. © OmniSci 2018 © OmniSci 2018
4. © OmniSci 2018 4 * open source for single node github.com/omnisci/mapd-core
5. © OmniSci 2018 NYC Taxis 1B Taxi Rides + 1M Buildings public demo: https://omnisci.com/demos/taxis/
6. © OmniSci 2018 6 Fast Hardware ( ) + Fast Software 3-Tier Memory Caching Query Compilation In-Situ Rendering
7. © OmniSci 2018 San Francisco Traffic Patterns 1B Traffic Points over 4 years + 71K Weather Points from 19 weather stations
8. © OmniSci 2018 © OmniSci 2018 • pems.dot.ca.gov California Department of Transportation (Caltrans) Performance Measurement System (PeMS) Data Clearinghouse • ncdc.noaa.gov National Climate Data Center • www.omnisci.com/blog Look for Abraham’s blog, with more detail, coming later today Traffic Patterns Links
9. © OmniSci 2018 © OmniSci 2018 • omnisci.com/demos Play with our live demos for yourself! • omnisci.cloud Get an OmniSci instance in 60 seconds • omnisci.com/platform/downloads/ Download a 30-day trial of OmniSci • community.omnisci.com Ask questions and share your experiences OmniSci Self Discovery Links
10. © OmniSci 2018 Thank you! Aaron Williams VP, Global Community @_arw_ aaron@omnisci.com /in/aaronwilliams/ /williamsaaron Abraham Duplaa Developer Advocate Consultant abraham.duplaa@cdtm.de /in//abraham-duplaa/ slides: https://speakerdeck.com/omnisci