You likely know what the Internet of Things (IoT) does. You’ve probably heard about some of its exciting implications both for your own industry and for the world at large. You may have learned that many other cutting-edge technologies, including artificial intelligence and the blockchain, will rely on the IoT to gather and process information quickly.
But there’s one thing about the IoT you may not have heard yet: without substantial changes to your network, it’s very unlikely to do much of anything for your enterprise.
Rather than a narrowly defined product or solution that serves a single purpose, the Internet of Things comprises a technological framework with a nearly infinite range of applications. You can’t just flip a switch and watch the results roll in. Earning an IoT-driven competitive advantage requires forethought, ingenuity, and a network architecture designed to support the IoT.
Accordingly, your network needs to be re-architectured to accommodate not just IoT technology, but your specific utilization of IoT technology before you can really unlock its full potential. Preparing for the IoT therefore entails two key steps: first, developing a vision for what exactly you hope to use it for, and second, how you can reconfigure your current enterprise network to make this application possible.
The promise of the IoT is a fairly broad one: to collect new streams of data by attaching sensors to everyday items, and use that data to enhance existing operations — and develop entirely new ones. With so many options available, determining which streams of data to collect, which devices to “connect”, and which data-driven optimizations to make is no simple task.
That’s why it’s critical to assess how IoT deployments can address the specific issues your enterprise is facing instead of just evaluating the IoT as a general whole. “Focus more on the applications, the customer, or the user experience rather than the technology,” advises Senior VP of CH Reynolds’ Data Division John Anderson. “Doing so will help you figure out where, what, and how you will need to build your network to accommodate current and future connectivity.” Pinpoint the issues that are hampering your business, and you’ll likely be able to imagine a way in which the IoT can solve them.
What do these bespoke IoT solutions look like in action? John Rossman, a former executive at Amazon, breaks down IoT use cases into three categories: those used to “improve customer experiences,” “create new business models,” and deliver “operational improvements internally.”
Amazon represents a great example of the first category, as its dominance of the retail business today is due in no small part to the degree of personalization it’s been able to achieve using massive volumes of customer data. Today, Amazon is expanding on that success by putting connected objects like Dash Buttons and virtual assistants (Alexa) into customers’ homes. These IoT-connected devices will make it far easier for consumers to make purchases, ask questions, and in the process, share information that enables Amazon to further optimize their customer experiences.
Perhaps the best known IoT use case is for driving operational efficiency. In a working environment like an oil rig, operational efficiency is about more than just keeping overhead costs down — a split-second delay in responding to an overheated pump could result in a dangerous situation for workers. Installing thermocouple IoT sensors on these pumps could virtually guarantee that they never exceed a given temperature threshold, thus eliminating equipment failures and protecting employees.
Finally, there’s the ambitious business of creating an entirely new business model using the Internet of Things. Here, a great example would be Tesla’s self-driving capability, whose “neural net” is built upon the IoT data collected from its connected vehicles. By using the collective driving experiences of all its current customers and transforming it into the entirely new product that is autonomous driving, Tesla has created entirely new revenue streams.
Organizations have to start with the outcome they want to make possible through the IoT, then work backwards to design a network configuration that makes that outcome possible. What changes you’ll need to make to your network will depend not just on your specific planned outcome, but on dozens of factors specific to your organization and its needs.
Why is it important to understand exactly how your organization will use the IoT? Because no two IoT applications will demand the same things of the enterprise networks tasked with running them. Oil rigs, for example, are typically based in remote locations without much wireless signal. Compute and storage resources in such situations are best located at the edge of the network so that IoT devices can function moment-to-moment without much latency. Meanwhile, Tesla’s self-driving system is being developed in tandem with an upcoming car-sharing platform (like Uber) that automatically changes each car’s settings to meet the preferences of the user. Since this all relies on individual driver profiles saved to a single cloud database, this network has to be organized in a more centralized fashion.
Enterprises that deploy IoT initiatives without a crystal clear organizational objective will end up with a solution to a problem they don’t have. IoT can’t solve every problem (not yet, anyway), and many challenges simply won’t have IoT solutions. But the successful IoT initiatives will start with an organizational need — say, for example, a computer chip factory needs to increase production on a specific assembly line.
Corresponding with that need, they’ll have a desired solution or outcome — for example, decreasing chip assembly time by 4%. From there, an IoT application is conceived — say adding sensors to robotic arms on the assembly line, aggregating that data in the cloud, and using machine learning algorithms to reduce minimum clearances by 50mm.
Then there’s designing a network architecture that can support the technology. Many IoT applications need real-time data, massive bandwidth, flawless coverage, and perfect reliability. The result is usually that networks need to be upgraded, and in some cases entirely redesigned to support IoT-enabled applications.
Implementation happens in reverse — first you build your IoT-optimized network, then you install your IoT tech, deploy your application, and (hopefully) achieve your desired results.
When it comes to preparing to launch an IoT initiative, enterprise IT pros will generally focus on three main factors: wireless connectivity, cybersecurity, and edge computing.
Among the most fundamental questions you must answer about your enterprise’s specific IoT deployment is how all your interconnected smart devices will actually “connect,” both to the internet and to one another. This decision is one that should be made carefully, as a miscalculation could translate into millions of dollars of waste.
Much of the current discussion about the IoT is centered around cellular technologies like 5G, LTE-M, and EC-GSM-IoT, for example, but these aren’t at all well-suited for transmitting small amounts of data infrequently. For many IoT implementations that revolve around regular “status checks” from sensors — like the thermocouple sensors from the oil rig discussed earlier, for example — this makes cellular-based communication tech the far less cost-efficient option.
Most IoT applications would benefit from instead using a technology like WiFi (especially the new IoT-friendly WiFi 6 standard). Factors like the total number of sensors, geographic coverage, data rates, and and end-to-end latency should be considered before deciding what you’ll use to connect your IoT network. That technology must also be implemented more or less seamlessly into your existing enterprise network, avoiding any interference or compatibility issues that might affect overall performance.
One of the unfortunate downsides of the rise of the Internet of Things is that the more endpoints exist on your network, the more opportunities hackers have to infiltrate it. IoT devices, whether they’re sensors, home appliances, or security cameras, represent attractive targets for cybercriminals because they provide anonymity in the form of an unsuspicious IP address. Having an IoT network means your company must be especially vigilant against cyber threats, which can hack your endpoints in order to steal information, participate in click-fraud schemes, obfuscate network activity, and far worse.
Part of this effort to bolster security should focus on basic measures like strong passwords. While Congress is considering legislation that would change this, only California has policies in place that force IoT device manufacturers to create non-default passwords (i.e. not “0000”) for each connected endpoint. Without strong passwords, your IoT devices could easily be incorporated into a malicious botnet.
Beyond that, IT teams will need full visibility into their IoT networks, using whatever software tools are necessary to understand exactly how many applications are being used on their networks. Enterprises should also perform a thorough network audit in order to establish activity baselines for all of its company’s IoT endpoints. This will make it easy for them to understand when a device is acting irregularly — by pinging other devices too frequently, for instance — and detect potential threats to their network early on.
Many of the game-changing applications of the IoT require the ability of smart endpoints to react almost instantaneously to incoming data. Edge computing makes all this possible by enabling enterprises to process and store data in the same place it’s created — at the “edge” of the network, rather than at the center. This significantly cuts down latency and makes it possible to make decisions in near-real-time. In many cases, edge computing is absolutely necessary in order to make IoT initiatives a practical reality.
However deploying an effective edge computing solution requires the rearrangement of your IT infrastructure. Relocating both the hardware and the fully redundant, uninterruptible power sources they run on to be closer to actual endpoints can be highly challenging in many cases. It also makes the work of monitoring and managing these now disparate endpoints more complex, often requiring that enterprises hire more IT personnel. Understanding the best way to make all this possible without the help of expert IT consultants can be intimidating, to say the least.
Any move to the Internet of Things motivated purely by industry hype is likely to cost more than it’s worth. The IoT could represent a significant opportunity for your business, to be sure, but without a great deal of clarity regarding exactly what you hope to accomplish, the costs of rearranging your IT network in order to support your IoT initiative will quickly add up. In other words, your IoT deployment has to not only make business sense, but IT sense, as well.
In an emerging arena like the Internet of Things, there’s nothing more valuable than having an experienced partner to help you ensure the technical side of your new endeavor goes off without a hitch. Turn-key Technologies (TTI) can be that partner.
At TTI, we have nearly three decades of experience helping enterprises design, deploy, secure, and manage high-performing wired and wireless networks. Our team of highly certified experts can help you reimagine your network architecture and bring greater connectivity to their workplace, whether that means moving your storage and compute infrastructure to the edge or simply ensuring your network has sufficient bandwidth to support all the new devices an IoT network entails.
As the business world races to create IoT applications that will disrupt their respective industries, those organizations that partner with networking experts like TTI will have the advantage of being sure that the technical side of their new venture is covered.
Reach out today to see how TTI can help you optimize your network for the IoT!
Please, rotate your device