Rethinking How Connected Vehicles Move Data

Background
and Motivation

Modern vehicles are becoming software-defined and increasingly AI-driven, and the volume of data they produce keeps growing — a connected vehicle can generate 1–2 terabytes of raw data each day. Sending everything via cellular networks in real time raises cost, congestion, and scalability concerns. What is needed instead is a way to choose the delivery path for each piece of data according to its urgency, value, and delivery deadline.

AECC’s answer is the Data-First Architecture: a communication and computation model that places data at the center of the ecosystem. Each piece of data is matched to the most suitable path based on its value and delivery deadline. Time-sensitive information, such as safety alerts, uses connectivity suited to timely and reliable delivery. Large, delay-tolerant data, such as training datasets and detailed logs, can be stored on the vehicle and offloaded opportunistically through Wi-Fi access points — or directly between nearby vehicles — when it is most efficient to do so. The result eases pressure on cellular networks while supporting faster, more sustainable data-driven development.

AECC’s Role

AECC brings automotive, network, and computing infrastructure stakeholders together to translate real vehicle-data requirements into shared architectural principles. Data First SIG develops scalable approaches for collecting, moving, processing, and storing data across vehicles, edge infrastructure, networks, and cloud platforms.

Data First SIG brings together experts from key industry stakeholders to define requirements, develop architectures, and validate them through publications and PoCs.