Large-Scale Data Transfer and Connected Vehicles
summary
Claims that connected vehicles generate ~4 TB/vehicle/day (Intel) while cloud interconnect sits around ~2 Pb/s (~8 ZB/year) and rising, creating a data gap where aggregate vehicle data far outstrips transfer capacity. Solutions store bulk sensor data in vehicle and near-edge, use three transfer patterns (application-specific, filesystem-level, volume replication), and must meet eight requirements: security, high-speed, fault-tolerant, scalable, manageable, automated, flexible and distributed controls plus cost-model-aware scheduling to optimize scarce bandwidth.
AI-written summary · about
chapters (15)
- 0:00Connected Vehicles and SOAFEE
- 0:44Data vs Cloud Interconnect Capacity
- 2:11Data Gap in Autonomous vehicle
- 3:08Automotive data requires edge storage
- 3:58Autonomous vehicle inbound bulk data transfer flow
- 4:47Three broad approaches exist for data transfer
- 6:15Common requirements for data transfer
- 6:448 fundamental data transfer requirements
- 7:42High-speed data transfer
- 8:28Fault-tolerant
- 9:10Scalable data transfer
- 10:00Data manageability
- 10:42Data automation
- 11:33Flexibility
- 12:27Distributed nature of data
related talks
24mHow to Drive Impact and Scale with Your Big Data/AI Applications?
37mThe global data plane for autonomous driving from Day 0 to Day 2 — Stephan Reimann and Simon Putz
06mData Driven Performance
24mCreating a Data Platform for Software-Defined Electric Vehicles
22mInvesting in Data Analytics
17mIt's All in the Cloud
33mCloud & data - The Engine & Fuel of Business
39mUnlock the potential of your data: Optimize your data store architecture...