Cloud IoT EdgeALPHA

Deliver Google AI capabilities at the edge.

Bring the intelligence Artwork

Bring the intelligence of Google Cloud to the edge

Cloud IoT Edge extends Google Cloud’s powerful data processing and machine learning to billions of edge devices, such as robotic arms, wind turbines, and oil rigs, so they can act on the data from their sensors in real time and predict outcomes locally. Cloud IoT Edge can run on Linux-based operating systems. Cloud IoT Edge is composed of two runtime components, Edge Connect and Edge ML, and also takes advantage of Google’s purpose-built hardware accelerator ASIC chip, Edge TPUTM.

Edge IoT Core artwork

Edge Connect

Securely connect edge devices to the cloud, enable software and firmware updates, and manage the exchange of data with Cloud IoT Core.

Edge ML artwork

Edge ML

Run ML inferences of pre-trained TensorFlow Lite models locally, significantly increasing the processing power and versatility of edge devices. This enables the next wave of machine learning applications and use cases.

Edge TPU artwork

Edge TPU

Google’s purpose-built AI chip designed to run TensorFlow Lite ML models at the edge. Edge TPU offers high performance per watt with a small footprint, enabling a broad set of use cases. This hardware accelerator complements Cloud TPU, which is used for training models in the cloud.

How Cloud IoT Edge works

IoT Edge Module artwork
Deploy ML models artwork

Deploy ML models at the edge for faster real-time prediction

Make real-time predictions for mission-critical IoT applications. Edge ML, built on a TensorFlow Lite runtime, uses CPUs and hardware accelerators like Edge TPUs and GPUs to run on-device machine learning models that provide faster predictions for critical IoT applications than general-purpose IoT gateways — all while ensuring data privacy and confidentiality. Plus, Edge ML and Edge TPU have been extensively tested to natively run open source benchmark models like MobileNets and Inception V3.

Enhance operational reliability artwork

Enhance operational reliability

Locally store, process, and derive intelligence from data at the edge, while seamlessly interoperating with the rest of the Cloud IoT platform. This allows you to build robust, enterprise-ready IoT solutions on premises, without having to worry about intermittent cloud connectivity. This can be critical for video and audio applications that require real-time processing and for serving offline use cases where devices can’t reliably connect to a network.

Secure your devices artwork

Secure your devices and your data

Securely connect your globally distributed devices on Google Cloud IoT platform. Cloud IoT Edge processes and analyzes images, videos, gestures, acoustics, and motion locally on edge devices instead of sending raw data to the cloud. This edge processing obviates the need to send large and potentially sensitive data streams to the cloud. Additionally, Cloud IoT Edge uses JSON Web Token to authenticate edge devices with Google Cloud instead of relying on the mutual authentication of a TLS stack.

Cost-effective artwork

Cost-effective at any scale

Optimize the integration cost of your IoT solution with a comprehensive software and hardware stack from Google Cloud IoT. This integrated stack uses Cloud IoT Edge, Linux OS, and Edge TPU to transform ordinary edge devices into smart and powerful gateways.

Industry use cases

Manufacturing_Icon@3x Created with Sketch.


Detect anomalies in high-velocity assembly lines that provide early warnings for potential issues, enabling you to get ahead of the problem with predictive maintenance.

Retail_Icon@3x Created with Sketch.


Deliver hyper-personalized product recommendations, offers, and communications based on in-store shopper behaviors.

Automotive_Icon@3x Created with Sketch.


Enable intelligent technologies like collision avoidance, traffic routing, and eyes off the road detection system.

Xee’s connected car platform makes driving simpler, safer, and more economical. Cloud IoT Edge and Edge TPU enables accelerated ML inferences at the edge, allowing the Xee platform to analyze images and radar data faster, detect potential driving hazards such as road conditions and tire wear and tear, and to alert drivers with real-time precision.

Romain Crunelle, Chief Technical Officer, Xee


Explore blogs and videos specific to machine learning and edge devices.

How to enable collaborative machine learning without centralized training data

On-device machine learning: TensorFlow on Android

Building a smarter world with machine learning and IoT data

Google Cloud

Get started

Cloud IoT Edge Alpha

Apply for early access to the Cloud IoT Edge and Edge TPU development board.

This product is in alpha. For more information on our product launch stages, see here.

이 페이지가 도움이 되었나요? 평가를 부탁드립니다.

다음에 대한 의견 보내기...