Scenarios

Internet of vehicles

The Internet of vehicles is a typical application of IoT technology in the intelligent transportation field. It uses technologies such as IoT, cloud computing, sensors, big data, and wireless communication to collect, store, manage, and analyze traffic information. These technologies enable traffic management, command and decision-making, and information services. This helps reduce urban traffic congestion and improve road safety.

Internet of vehicles scenarios involve many vehicles, large data volumes, various data types, and diverse data sources. Therefore, these scenarios require support for high-concurrency writes, storage, and statistical analysis of massive amounts of vehicle data. The core requirements for data storage and analysis in these scenarios are as follows:

• High-concurrency data writes: Support real-time writes from millions of nodes for many vehicles.

• Data storage: Store vehicle trajectory and status information for long periods. This requires support for extremely large single tables and cost-effective storage.

• Real-time geo-fencing: Generate geo-fence alerts based on real-time data. This requires integration with real-time computing for real-time statistical analysis.

• Massive vehicle management: Manage, query, and analyze status information for a massive number of vehicles. This requires integration with batch computing to query and analyze large volumes of historical data.

The following figure shows the architecture for the Internet of vehicles. Vehicle data is ingested into a Kafka message queue using the Message Queuing Telemetry Transport (MQTT) protocol. A portion of the data is then sent to Flink for real-time computing, such as traffic scrubbing and geo-fence checks. The remaining data is written directly to Tablestore to store vehicle information. The system also supports generating reports using the SQL protocol for data statistics and connecting to MaxCompute for large-scale historical data analytics.

Smart home

A smart home uses technologies such as automatic control and IoT to centrally manage smart household devices, such as refrigerators, washing machines, air conditioners, TVs, and speakers. This enables features such as appliance control and home environment monitoring. It provides users with an energy-efficient, convenient, safe, and comfortable personalized living environment.

The core requirements for data storage and analysis in smart home scenarios are as follows:

• High-concurrency data writes: Support real-time writes from tens of millions of nodes for many devices.

• Data storage: Store collected business data for long periods. This requires support for extremely large single tables and cost-effective storage.

• Data analytics: Analyze device usage data to promptly detect anomalies and enable control. This requires integration with a compute engine for large-scale data query and analysis.

Industrial IoT

In the industrial realm, technologies such as IoT, sensory control, information processing, and network communication are applied to all stages of the industrial production process. This improves production efficiency, enhances product quality, and reduces resource consumption.

In an industrial production process, devices typically report their latest operational status at fixed intervals or based on event triggers. This information includes the device ID, current operating temperature, humidity, and pressure values. Users can manage devices based on their metadata. For example, they can query the current status of a specific device, perform online searches for devices using multiple conditions, or select a collection of devices that meet certain criteria. By analyzing device metadata, users can monitor the operational status of devices in real-time, respond promptly to anomalies, and prevent failures.