Data Efficiency and MQTT: Ways this Lightweight Protocol Can save on Cellular Costs Today
1.Queuing: Communication limited to a small code size allows the m2m client to send published values (data points)collected from the m2m server device(such as a sensor)to the cloud-computing broker, where any access or further data processing can be sent to local area networks unconstrained by data restrictions; the m2m part of the telemetry compresses and rations the packets sent out, whereas the machine to cloud, and machine to human portions of the Pub-Sub (publish and subscribe) telemetry can relay back and forth more freely.
2.Actionable Messages: Actionable data sent over MQTT, as with digital or analog I/O for instance,saves enough money to give the operators the choice between using a cell budget to access and reconfigure the setpoints of field bus clients remotely, through VPN for instance, or sending personnel to deal with the remote site if remote access fails to resolve the problem.
3.Telemetry Prioritization: GPS is a data efficient technology when the positioning results are communicated to a networked device over MQTT. If your remote monitoring is on the move, as with a fleet management application, then it’s imperative that positioning data be of the right level of importance, and that it be sent in such a way that keeps cellular data consumption costs low–as there’s no doubt that mobile GSM expenses can hurt a company’s bottom line, especially where less than critical information must consume more bandwidth to reach the MQTT Broker. MQTT as a protocol makes assessments of the event-data’s urgency, and queues events’data for efficient transport.
4.Alarm Conditions Only: With a fieldbus setup like Modbus, polling events communicable over MQTT makes SCADA a low-cost proposition. Rather than sending Modbus read-commands over cellular to the supervisory level client, the Modbus control level client can queue alarm condition data and publish events to the MQTT broker. From the broker, the supervisory level(using devices such as personal computers, smartphones, and tablets)can be notified of subscribed-to topics and can publish new setpoints back to the Modbus client via the cloud broker.
5.Computing on the Edge: At a remote site IIoT integrated via MQTT, multiple sensors and (or) data processing units can communicate with one another, and if necessary, can be formed into a distributed cloud-like processing network that serves as a local extension of the cloud –in essence, fog computing or edge computing. A fog will be able to roll down from the cloud to the remote site, and much of the data processing work that would be done in the cloud (distributed processing over the internet) can be done right there at the remote site; rolling the fog down to your remote site would mean that only the most essential data would be published to client devices via MQTT Broker, which in turn would mean that cellular data consumption would be kept to a minimum.