Connecting an OEM device to IBM Watson IoT Platform

IBM Watson, a next generation supercomputer powered by over ninety(90) servers processing at a mind-boggling rate of 80 teraflops by IBM seems to be the future of computing. Watson, named after the founder of IBM, Thomas J. Watson has been the forefront of the development of connected devices or simply – Internet of Things.

IBM Watson under the hood:

  • Computation Speed: 80 teraflops
  • Concurrent Servers: 90
  • Pages of Information: 200 million pages
  • No of logic rules: 6 million
  • No of Processor Cores: 2880
  • RAM: 15TB


Ever since 2015, the world is abuzz with the concept of devices continuously talking to each, sharing information and arriving at decisions without the need of human intervention. real-time in simple words is Internet of Things along with Machine to Machine Communication (M2M). IoT has since then been ever growing exponentially and is predicted to be valued at $ 7.1 trillion by connecting over 30 billion devices by the year 2020.

With the advent of various industrial and home automation systems, the smart devices penetrated deep into our daily life right from getting real-time traffic updates and route suggestion while leaving for work to getting filtered recommendations on Netflix, YouTube by analyzing user behavior patterns. IBM’s revolutionary IoT platform Blue.

As a developer, the rule of the thumb is “Don’t reinvent the wheel till the time it isn’t broken” and IBM seems to have followed the same principle while building the Watson. Watson is a one stop solution for all of your computing needs. Right from securely storing the data to running analytics and getting real time insights to further improve your business process.

The first step in connecting your existing OEM devices to Watson IoT platform is collecting the data from all the sensors and passing it through to the Bluemix server via lightweight messaging protocols like MQTT in a secure way. Once the data is onto the IBM servers you need not worry even a bit about the security. Next, you could either build your own or use existing Machine Learning models provided by the IBM Watson IoT platform. You could soon get real-time analytics, Key Performance Indicators (KPI) to understand the data in a meaningful way.

As a developer, the main advantage of the IBM Watson platform is not worrying about the underlying tasks like creating, maintaining and securing servers every now and then to store the data, building Machine Learning models, training them and testing them and tweaking for maximum efficiency. Also building and maintaining an edge analytics system for real time analytics. All these things come packed with the Watson IoT platform along with the reliability of IBM.

Recently, we had a chance to work with a startup working in the IoT sector providing end to end solutions right from hardware to software. They had quite a bloated team consisting of two DevOps engineers for maintaining the integrity of the servers, four Data Scientists to build and train Machine Learning models as well as the Analytics Engine, an integrations expert to get the data from the hardware to the platform as well as a person to visualize the data and also a pentester to keep the environment safe. This is quite a bloated system for a startup at a very nascent stage with not much of funding in hand. The issue they were facing was bringing their product from idea stage to the real world. While we suggested them to move on their platform to IBM Bluemix they had a staggering 210% growth in productivity and their product was out on the market within 36 days. Also, their team was reduced by more than 60% with most of the work being integrated directly onto the platform.


Use Cases

The advantage of using a platform life Watson is the ability to reuse the dataset already compiled and used to train the model. This lets you to directly start using it without any overhead time in creating and training models. All you require to start off is a few API calls.

Earlier on we used to write code based on certain pre-defined rules, but as the amount of data increased along with the processing power of devices, we moved on to a more cognitive based computing. IBM’s core is cognitive computing that gives it a human touch to computing. Watson is constantly trained by making it consume more and more data every day and learn from it.

A startup based out of the USA is currently working on a system which guides doctors on the best possible treatment for a patient based on their age, gender, medical history, etc thereby drastically improving the healthcare facilities around the world. This is possible because of the vast library of medical knowledge amassed by Watson under the guidance of experts in medicine from all over the world.



IBM’s platform is a must use for budding startups working with minimal resources as it helps reduce the turnaround time drastically by using the inbuilt models of Watson. With Watson, you could simply focus on building your product and leave everything in the reliable hands of IBM. Another main feature I loved as a developer was high scalability when required. Whether our client base is 30 or 300 IBM’s flexible servers can adapt to our requirements anytime.



Feel Free to share your thoughts in the comment section below.
Don't forget to follow us on Twitter, like our Facebook Fan Page and Add us to your circles on Google+ to keep you updated with the latest technology news, gadget reviews, launches around the world and much more
Tagged with