Engineers looking to add Internet of Things (IoT) functionality to their industrial systems will soon discover that there are a lot of cloud platform options to choose from. These cloud platforms offer the computational and storage backbone for the Industrial IoT (IIoT). As a result, this decision will impact the very core of an IoT implementation.
In other words, industry specialists need to start shopping around for an IoT cloud Platform now or risk ending up behind the eight ball.
But this leads to another problem. With this IoT development boom, Gartner also predicts a boom in IoT Platform as a Service (PaaS), which is a fundamental part of an IoT cloud Platform. Gartner said that by 2020, more than half the applications developed on a PaaS will be centered on the IoT.
In other words, the IoT device and Cloud Platform booms further complicate the selection of the right tool for the job.
Cloud Computing Tools to Look for in a Cloud Platform
So, what does an IoT cloud platform do anyway?
“The Cloud is the most important trend shaping the future of the smart, connected world,” wrote Brian Krzanich, CEO of Intel. “It means that everything that a ‘thing’ does can be captured as a piece of data, measured in real-time and is accessible from anywhere.”
When breaking down the computational aspect of these tools, there are three basic functions: infrastructure as a service (IaaS), PaaS and software as a service (SaaS). This doesn’t even include all the security services you will need to have embedded in your cloud platform to protect it from intrusion.
IaaS is the connective tissue that keeps an IoT system together. This includes the provision processing, storage, networks and basic computational capabilities. In effect, it is what makes the deployment of PaaS, SaaS and other software possible.
“Customers don't manage or control the IaaS, but they are able to access the servers and storage, and can operate a virtual data center in the cloud. There are a lot of IaaS providers,” wrote Randall Scasny, “The biggest ones are Amazon Web Services (AWS), Windows Azure, Google Compute Engine, Rackspace Open Cloud, and IBM SmartCloud Enterprise. GE’s Predix platform for the Industrial Internet will be available to run on Microsoft Windows Azure Cloud.”
The PaaS is used to run, develop and test applications on the device over the cloud. Scasny explained that the engineers also don’t have much control over the infrastructure of the PaaS. However, they can control which applications will run on it. These engineers will also be able to control the configuration of the app hosting environment.
Some options for an industrial PaaS include:
- GE’s Predix
- Honeywell Sentience
- Siemens's MindSphere
- Cumulocity
- Bosch IoT
- Carriots
The final piece of the IoT cloud computing puzzle is SaaS. This system gives users access to applications that control their IoT system on a web-browser.
“The advantage of a SaaS is that you don't need to run or install specific applications on individual computers, which not only saves time and money, it also simplifies maintenance and support,” wrote Scasny. “Common SaaS examples include: Google Apps and Cisco WebEx. Siemens's Industrial Machinery Catalyst on the Cloud is an example of an Industrial SaaS [which uses the] AWS infrastructure.”
Comparing IoT Cloud Platforms: Consumer vs. Industrial
The major difference between a consumer IoT and an IIoT cloud platform will be on the focus.
IIoT platforms, however, will talk to the engineer running an industrial system. They will focus more on operational and predictive maintenance technology.
For instance, Scasny noted that “[IIoT platforms] are designed to allow data gathering throughout manufacturing production processes, in order to improve performance as well as predict failures before they happen.”
As a result, if you are creating an IoT system for an industrial setting, you can likely cut many IoT cloud platform options from your shortlist.
Functions to Look for in an Industrial IoT Cloud Platform
The two aspects of an IoT cloud platform that should be at the forefront of any implementation is data security and data integrity. “Cybersecurity is a cat and mouse game where the mouse gets bigger and more ferocious by the minute,” said security analyst Joshua Greenbaum. “Security threats can’t be minimized. That’s why companies of all sizes are running to the cloud.”
The thought process here is that the bigger and more reputable the cloud service is, the more they can afford the top cybersecurity talent. This logic can also be applied to data integrity; the bigger cloud services will be able to afford a computer engineering team that can ensure the data stored on their cloud will be the same as the day it was recorded. Therefore, by picking an IoT cloud platform with top-notch security and data integrity teams, your organization can focus on the security and integrity threats specific to your designs and IoT system.
However, it is important to note that choosing an IoT cloud platform with the top security option doesn’t mean that your IoT engineering team can sit back and relax on security. “From a design point of view, engineers need to learn about hacking security. You need security at the edge point to make an intelligent analytic device,” said Michael Wendenburg, CEO at Michael Wendendenburg Online Redaktion. “If you hack into that point, you hack into all this data. Engineers are not prepared for that.”
Another common issue that engineers will find after implementing an IoT system is that the data they collect just sits there. In fact, IBM reports that a whopping 90 percent of data collected by connected devices will remain unused.
The problem with this is two-fold. First, there is the sheer volume of data being generated and collected. Second, there is the analysis of this data to turn it into something meaningful. The volume of data becomes particularly complicated when designing an IIoT system.
“[Don’t underestimate] the volume of data in an industrial system,” warned Scasny. “Some of these industrial systems will have huge sensor networks, so the platform must be scalable to handle very large volumes of data.”
The sheer volume of data from different pieces of industrial equipment will also lead to data compatibility issues. Each data set might have a different volume, sampling rate, chronometer and more. Any industrial IoT cloud platform worth its salt will need to be able to handle these complications.
However, data volume isn’t the only roadblock causing IoT information to sit idle. After all, data is only as important as the analytics you can squeeze out of it. As a result, it is also important to choose a cloud platform that has a focus on Big Data analytics.
“The data coming from connected devices and sensors can be endless, but it is often untapped or underutilized in terms of its potential impact to their business,” said Sam George, partner director of Azure Internet of Things. “When you are able to combine and analyze device and asset data with other types of business data, you may be able to uncover insights that were out of reach in the past.”
“There is much more value to unlock from the cloud and data center, and analytics are the key to that,” agreed Krzanich.
But it isn’t enough to just have your IoT systems crunch data; they should be able to react to the data and package it into a format the engineering team can visualize.
There are IoT cloud platforms that will give engineers the ability to set up triggers in the event that data analytic algorithms reveal certain outcomes. These triggers can set off warnings, run applications, shut down equipment or optimize a system’s performance. As a result, engineers should really look into the analytical capabilities of the IoT cloud platform of their choice.
As for data visualization, this is typically done using dashboards. Choosing an IoT cloud platform that gives the IoT design team control over the dashboards is vital; pre-built, alone-size-fits-all dashboards won’t cut it.
It’s even possible a different dashboard might be needed for each product variation and user role. Make sure the right tools are available so that these dashboards are customizable and only viewable by the appropriate user role.
Another key factor in choosing an IoT cloud platform is interoperability. With everyone inventing IoT products it’s no wonder that the communications between connected devices has become a Tower of Babel.
Engineers must therefore ensure that the equipment they wish to connect will be compatible with the platform they choose. Since IoT engineers will often be connecting existing equipment, each with a separate application program interface (API), this can be a serious sticking point.
“As more and more consumer IoT devices are created and brought to market, the need for interoperability becomes paramount, as it is the key to creating more compelling and integrated experiences,” said BK Yoon, CEO of Samsung, which created the ARTIK Cloud. “The need for an open cloud solution that can work with any connected device and with other cloud services is critical for broader consumer adoption.”
In other words, if an IoT implementation will need to call on multiple cloud systems, then choosing a “cloud agnostic” IoT platform gives IoT engineers a particular advantage.
“Our open IoT platform strategy is driven by our customers’ need to grow and scale their IoT projects,” said Jim Heppelmann, PTC president and CEO. “By working closely with IoT cloud leaders, we can unlock unprecedented value from the IoT, which will further enhance our customers’ products and offerings.”
For additional information, see “Will There Be A Dominant IIoT Cloud Platform?” by Randall Scasny
Premier Farnell PLC has sponsored this post. It has provided no editorial input.