IoT service platforms – also called IoT Platform as a Service (PaaS) – are on the uprise and offer companies the possibilities to save costs. They also improve services and capabilities of products and shorten time-to-market. In the following we will take a look at the challenges of IoT platforms and how IoT PaaS can help to overcome them.
The IoT will be realised through a great number of systems, which will have to fulfil different requirements. Those will make use of hardware and software components, which will vary greatly according to specific needs and goals. If we take a look at the connected objects and end-nodes within the IoT framework and isolate them, they will be very similar to “common” stand alone products. Thus, the big difference of IoT systems and products lies in the backbone with logic and service-oriented features. Implementation of such features can be a complex and difficult task especially if scalability, flexibility and system simplicity is desired.
In order to tackle this challenge companies can invest in IoT PaaS to get access to IoT platforms.
The IoT creates systems and sub-systems through the inter-connection of devices in order to exchange, analyse and make use of data. Therefore changes, updates and failures of single components potentially affect a number of systems and users. In contrast to IoT products former stand-alone devices were easier to manage, as modifications usually did not affect other systems and services.
This new and different behaviour of sub-systems and systems, as well as their linkage, increases complexity of IoT products.
In order to be able to manage a big number of IoT devices APIs are important. They enable the update and exchange of hardware and software in levels below them without affecting the system above. Those APIs, as well as high-level logical architecture and system layout, templates are usually part of IoT service platforms and free up time for designers and developers to concentrate on product design and customer needs.
The product development will be altered by big differences regarding product lifecycles of IoT system components. Commonly, product roadmaps have been predictable and all steps along the way from inception, definition, development, test, deployment, volume ramp to end of life have been timed in advance. In an interconnected system with distributed components this is no longer possible.
Sensors and actuators within the system might be in place for a long time (years and even decades) without changing. Other components however might change with high frequency and therefore the corresponding API is the only constant. Furthermore, several products will change at unpredictable frequencies.
In addition, not all component failure modes can be anticipated. Components might be added or subtracted randomly to an IoT platform and product lifecycles will not synchronise. Consequently, it is crucial that a service platform is flexible, detects failure, routes around malfunction and offers scalability.
All endpoints are connected to a larger solution and represent only a part of the physical and logical design of the IoT platform. Thus, the development of IoT products needs to include additional networking components.
As soon as the IoT platform has been set up, the endpoints collect data and send it back to the R&D where it is analysed. Product redesigns are not tied to a pre-defined time schedule anymore. Instead upgrades and replacements are developed and carried out based on continuous evaluation of data and new requirements for the single components as well the complete system.
Let’s take a look at a possible use case. A manufacturer of thermostats that wants to evolve to be a provider of smart home devices will need to create and manage an IoT platform. A platform that includes own devices and endpoints like connected thermostats and temperature sensors. In addition the customers’ smartphones will be part of the system and functions as remote control. Other devices from other manufacturers might be part of the customers’ home set up and influence the IoT platform requirements. Cloud connectivity and data analysis along with the necessary hardware and software complete the system.
Development of the thermostats and sensors as well as a smartphone app might be a reasonable task for the engineers of the company in the example. However they may lack the knowledge and experience necessary to develop Do-it-Yourself (DIY) IoT platforms for their products.
In case it is possible to develop a DIY platform leveraging existing technical resources and technologies it might still be difficult to constantly manage such a system.
Investing in an IoT platform service can be a good alternative. IoT PaaS offers companies the ability to select services and proven APIs according to their requirements and quality standards. This will help to reduce time-to-market significantly. Outsourcing the development of the platform further contributes to free up capacities, which are necessary to optimise and create new products and services based on crowdsourced IoT data.
In conclusion it seems that developing DIY IoT platforms might be challenging to cope with for smaller and medium sized businesses. This is especially the case if they do not obtain experience with network and cloud technologies.
IoT PaaS solutions are a good way to address common challenges during the process of moving towards the IoT. Additionally, those services help to predict and optimise IoT product related costs like operational expenses requirements of your customers.
At EBV we are able to support your products not only with hardware. We can connect you to partners like Ayla Networks offering sophisticated IoT service platform solutions.