CarVi Meetup - Automated Vehicle Telematics with OmniSci

Transcript

1. Hardware-Acceleration of Automated Vehicle Telematics CarVi Meetup | San Francisco | April 30, 2019
2. Veda Shankar Senior Developer Advocate, OmniSci Community veda.shankar@omnisci.com slides: https://speakerdeck.com/omnisci Dellaena Maliszewski OSS Community Manager, OmniSci Community dellaena.maliszewski@omnisci.com
3. © OmniSci 2018 © OmniSci 2018
4. © OmniSci 2018 Telematics Challenges said the biggest challenge is data utilization 48% * “2016 Telematics Patterns, Trends and Challenges” http://fleetanswers.com/sites/default/files/2016%20Telematics%20Survey%20Report_0.pdf
5. © OmniSci 2018 Telematics Challenges Majority of respondents only occasionally use telematics data for decision making * “2016 Telematics Patterns, Trends and Challenges” http://fleetanswers.com/sites/default/files/2016%20Telematics%20Survey%20Report_0.pdf
6. V. A. S. T. VOLUME | AGILITY | SPATIO-TEMPORAL
7. The Fastest Software Designed for the Fastest Hardware HARNESS GPUs
8. © OmniSci 2018 8 * open source for single node github.com/omnisci/mapd-core
9. © OmniSci 2018 NYC Taxis 1B Taxi Rides + 1M Buildings public demo: https://omnisci.com/demos/taxis/
10. © OmniSci 2018 10 Core Density Makes a Huge Difference
11. © OmniSci 2018 11 SSD or NVRAM STORAGE (L3) 250GB to 20TB 1-2 GB/sec CPU RAM (L2) 32GB to 3TB 70-120 GB/sec GPU RAM (L1) 24GB to 256GB 1000-6000 GB/sec Hot Data Speedup = 1500x to 5000x Over Cold Data Warm Data Speedup = 35x to 120x Over Cold Data Cold Data COMPUTE LAYER STORAGE LAYER Data Lake/Data Warehouse/System Of Record Advanced Memory Management
12. © OmniSci 2018 12 Fast Hardware ( ) + Fast Software 3-Tier Memory Caching Query Compilation In-Situ Rendering
13. © OmniSci 2018 Three Ways to Get Started GitHub repo OPEN SOURCE OmniSci as a service OMNISCI CLOUD Contact sales ENTERPRISE 13
14. © OmniSci 2018 Vehicle Telematics Analysis for Carmakers Analytics Challenges Scale and speed of telematics data overwhelms mainstream platforms Data-driven collaboration across the supply chain is difficult “The Last Yard” of tech adoption blocks desired business insights OmniSci Solutions Spatiotemporal analysis at a scale to capture insights from millions of cars OmniSci’s open source core eases data integration Ease of adoption speeds broad uptake, for rapid ROI 14
15. © OmniSci 2018 “We use OmniSci on-the-fly to investigate all the features and how they affect the output. We’re doing that with over 500 million points.” DR. ZACH IZHAM, PROJECT MANAGER, VW DATA:LAB
16. © OmniSci 2018 F1 Racing Demo Real-time Vehicle Telematics blog post: https://www.omnisci.com/blog/collecting-telematics-data-from-the-omniSci-grand-prix
17. © OmniSci 2018 F1 Racing Demo Real-time Vehicle Telematics Using Open Source Tools public repo: https://github.com/omnisci/vehicle-telematics-analytics-demo
18. © OmniSci 2018
19. © OmniSci 2018 Step 1: Data Engineering and ETL 3 pipelines • UDP to Kafka • Parse to JSON • Data Refinement (and insertion into OmniSci using JDBC)
20. © OmniSci 2018 Step 2: Querying and Visualization Plotly Dash using pymapd • Queries every 5-15 seconds • No indexing means the data is instantly available
21. © OmniSci 2018 Step 3: Create Your Own s3://mapd-cloud/DataSets/vehicle_telematics_dataset_f12018/ We’ve made the data available for the community Please copy the data rather than direct linking
22. © 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
23. THANK YOU Ask questions and share your experiences @ https://community.omnisci.com