Through all this, we here at Google Cloud are excited to help make this world a reality. We recently published a solution guide that describes how various Google Cloud Platform (GCP) services fit into the picture.
A data deluge
Vehicles can produce upwards of 560 GB data per vehicle, per day. This deluge of data represents both incredible opportunities and daunting challenges for the platforms that connect and manage vehicle data, including:
- Device management. Connecting devices to any platform requires authentication, authorization, the ability to push update software, configuration and monitoring. These services must be able to scale to millions of devices and constant availability.
- Data ingestion. Messages must be reliably received, processed and stored.
- Data analytics. Complex analysis of time-series data generated from devices must be used to gain insights into event, tolerances, trends and possible failures.
- Applications. Business-level application logic must be developed and integrated with existing data sources that may come from a third party or exist in on-premise data centers.
- Predictive models. In order to predict business-level outcomes, predictive models based on current and historical data must be developed.
GCP services, including the recently launched Cloud IoT Core provides a robust computing platform that takes advantage of Google’s end-to-end security model. Let’s take a look at how we can implement a connected vehicle platform using Google Cloud services.
|(click to enlarge)|
Device Management: To handle secure device management and communications, Cloud IoT Core makes it easy for you to securely connect your globally distributed devices to GCP and centrally manage them. IoT Core Device Manager provides authentication and authorization, while IoT Core Protocol Bridge enables the messaging between the vehicles and the platform.
Data Ingestion: Cloud Pub/Sub provides a scalable data ingestion point that can handle large data volumes generated by vehicles sending GPS location, engine RPM or images. Cloud BigTable’s scalable storage services are well-suited for time series data storage and analytics.
Data Analytics: Cloud Dataflow can process data pipelines that combine the vehicle device data with corporate vehicle and customer data, then store the combined data in BigQuery. BigQuery provides a powerful analytics engine as-a-service and integrates with common visualization tools such as Tableau, Looker and Qlik.
Applications: Compute Engine, Container Engine and App Engine all provide computing components for a connected vehicle platform. Compute Engine offers a range of different machine types that make it an ideal service for any third-party integration components. Container Engine runs and manages containers, which provide a high degree of flexibility and scalability thanks to their microservices architecture. Finally, App Engine is a scalable serverless platform ideal for consumer mobile and web application frontend services.
Predictive Models: TensorFlow and Cloud Machine Learning Engine provide a sophisticated modeling framework and scalable execution environment. TensorFlow provides the framework to develop custom deep neural network models and is optimized for performance, flexibility and scale — all of which are critical when leveraging IoT-generated data. Machine Learning Engine provides a scalable environment to train TensorFlow models using specialized Google computing infrastructure hardware including GPUs and TPUs.
Vehicles are becoming sophisticated IoT devices with built-in mobile technology platforms to which third parties can connect and offer advanced services. GCP provides a secure, robust and scalable platform to connect IoT devices ranging from sophisticated head units to simple, low-powered sensors. You can learn more about the next generation of connected vehicles with GCP by reading the solution paper: Designing a Connected Vehicle Platform on Cloud IoT Core.