Eat your own dog food is one of the most heard words from Sridhar Vembu when I was working for Zoho. This is a practice done while we were creating Zoho Writer and Zoho document management system and is practiced today as well. It is nothing but using the software created by your company for your customers for their own internal purpose. This is one of the most simple and powerful ways to understand the problems faced by your customer. It helps in identifying enhancements you can make on your software. We transformed the MQTT Broker into an IoT Application Framework practicing the same. Over the past few months, we were building products for manufacturing execution system above our MQTTRoute
Data Collection & Storage
Against all the hosted IoT Code vendors like Google, AWS, Azure and the other MQTT Brokers like HiveMQ , Mosquitto, from the initial days, we wanted to build something more than just being a broker. Message brokers just transacts messages between interested machines & devices. One of the most not so scalable solution we inferred in the market is building a manager application behind an MQTT client. MQTTRoute had options to store data into any database which can be processed by the manager application. We had multiple hooks to receive data and manage your edge devices by sending commands to the edge devices. The MQTT Broker came with an user interface ( a more technical one ) to view and manage devices.
What we learnt?
We have been seeing some of our customers directly querying our database or integrating it with the Google PubSub or Redis or Tableau for further processing and visualizing data. This was happening in spite of our user interface and device management options like send command to devices and the rules engine. However, MQTTRoute was doing the functions it was intended for flawlessly.
We started building the Daily production monitor for manufacturing industry ( discrete process) as part of one of MES initiatives. The development cycle helped us understand that even though the broker was doing its core right, But it does not 100% cover the requirements for building up a complete industry 4.0 application needed without a parallel process running.
MQTTRoute 2.0 — The complete IoT Application Framework
We have been working hard to ensure that the complete internet of things application including user interface customization, data aggregation & analysis, event data comparison with the processed data. All custom implementation can be done using the additional hooks. These hooks as of now are python based. We expect the same to be rolled out in a week’s time. We believe the new IoT application framework will help building and managing the industrial IoT applications faster and much easier within a single process.
Feel free to write to us if you have any specific need that needs to be addressed for your Industry Implementation.
Originally published at https://www.bevywise.com on July 23, 2019.