Will enterprise IoT become BYOD on steroids?

If you thought BYOD was a pain the neck for IT management, strap yourself in.

The integration of Internet of Things (IoT) devices in the enterprise is moving beyond the hype into a much more pragmatic and operational phase of planning, execution and service delivery.

There’s massive interest and buzz around IoT within companies of every type. This is largely due to the transformative business impact IoT will have. Major business will be transformed, disrupted or out completely when IoT is properly used in the business transformation process.

But with IoT solutions comes new demands and a wide range of technologies, many of which will take the enterprise into unfamiliar territory, requiring different analytical skills and management tools. In other words, IoT devices quickly finding their way into to the enterprise will make life a living hell for network managers – far beyond the problems they experienced with BYOD.

Unlike BYOD, IoT tools are “headless,” typically tied to line of business to drive top line revenue or bottom line cost cutting objectives. This means the importance of monitoring and managing of these new things, to ensure the best possible performance over computer networks, will eclipse that of conventional networked clients.

Managing IoT performance on enterprise networks

With all the power and benefits of IoT, IoT will also present a new host of challenges to enterprise IT teams that will exceed other recent challenges enterprise IT teams have had to deal with like interoperability, protocols and security.

IoT management is further complicated by the fact that some IoT devices have limited hardware capabilities, restricted networking capabilities and don’t run operating systems that support conventional IT or mobile device management. What’s more, IoT management tasks may be split across different factions in IT or network operations. Without a single source of insight into the performance of IoT devices that can be used by all the different networking constituents, more finger pointing among IT staff is sure to result.

Another difficult thing for network managers to get a grip on is the impact of IoT-networked devices on capacity planning. What impact will deploying 1,000 HD IP video cameras have on the access network, user performance or WAN links?

Some vendors are entering the IoT management space from adjacent spaces such as mobile management, so their technology may not be well-suited to all IoT devices or management tasks. Others, such as those offering new network analytics solutions are. These approaches are purpose-built for the task and much better suited for sorting out some of these problems. 

Who’s deploying what?

Manufacturers are deploying IoT robots on the production line. Retailers are using more sophisticated wireless scanners for inventory control and healthcare organizations are now integrating bedside telemetry monitors to streamline patient care, video cameras for security and android-based Wi-Fi phones for secure voice and text communications between staff and clinicians.

However, many of these headless devices often use specialized or proprietary protocols that behave differently with applications and network services than conventional laptops or smart devices on the network. Here’s a couple of real world examples:

Popular GE bedside monitors that track vital patient conditions send waveform data over the network using a proprietary protocol. Watching how this protocol behaves on the network and integrating the waveform data into useful insights that the IT team can actually use isn’t something that traditional network management and monitoring systems can do.

Another painful example is the use of emerging Wi-Fi communications solutions. Ascom android-based MYCO Wi-Fi phones have become wildly popular for voice and data communications. While network managers can monitor the connectivity of these phones, they can’t easily determine the user experience or call quality easily as they are used across the infrastructure. Each phone sends an uncommon SYSLOG message format to backend servers with all the useful details to get a clearer picture of the device performance. These SYSLOG messages must be parsed, analyzed and correlated with the other device transactions across the network to be able to pinpoint where call problems occur.

Taking a broader approach to IoT device performance management

Managing the IoT device performance on the network requires a broader approach that encompasses a more complete understanding of how these devices interact with other parts of the network, network services and applications.

Conventional infrastructure management tools were never designed to analyze each device network transaction across the full stack. Consequently, if any device can’t get an IP address, reach an application or gets sticky to the least favorable Wi-Fi access point, business suffers and IT staff must go hunting.

New network analytics platforms are being developed for this precise purpose.

These systems are able to ingest virtually any kind of data running over the network and, using advanced analytics and machine learning techniques, determine how each IOT device transaction is performing with every part of the network. Unlike traditional management and monitoring solutions, these emerging platform focus on device performance leveraging more modern technologies such as machine learning, cloud computing and artificial intelligence. Rather than merely displaying raw data in charts and graphs or visual displays of different types of data, these systems constantly measure and correlate real user, network, application and device data across all layers of the network and translate it into plain English actions that IT staff can use for more proactive remediation.

This is where the over used phrase “user experience” becomes immensely important. Despite these devices not having any user operating them, understanding, tracking and managing their performance on the enterprise access network can be easily equated to tangible business value, as opposed to user productivity which tends to be much softer and harder to quantify in real monetary terms.

Ultimately network managers will need to take a fresh look at how to best manage the performance of new IoT devices on their network or run the risk of losing the real value these investments can bring the business.