When you think of the "Internet of Things" (IoT), the first things that come to mind may be the wearable fitness tracker or the smart TV you use to stream your favorite shows. Indeed, IoT is one of the hottest consumer trends, and expectations are that more than 20% of US consumers will own smart refrigerators, smart watches and other connected devices by 2020.

While IoT is gaining traction with consumers, it's already having a transformative effect on the business world. Companies can analyze the volumes of data connected devices supply to improve decision-making processes and to help employees be more productive. However, as an enterprise brings more connected devices online, IT will struggle to maintain adequate application performance levels. Implementing application performance monitoring (APM) establishes the end-to-end visibility IT needs in order to immediately identify what's causing an application to perform poorly, so that they can fix the issue before users complain.

More enterprises are embracing cloud computing and creating hybrid IT architectures comprised of a mix of on premise and cloud-based applications. The move to the cloud is driving the adoption of connected devices, and there are no signs of slowing. Cisco Systems reports that in 2008 there were already more things connected to the Internet than people. By 2020, the amount of Internet-connected things will reach 50 billion.

IoT already powers a wide range of business applications. For example, oil and gas companies can track oil flow through pipelines, smart meters help measure and optimize the delivery and consumption of electricity, and mining companies can better monitor logistics. A Dutch company even uses Internet-connected sensors on cattle to alert farmers when the animals are sick or pregnant.

Just as the number of business use cases for connected devices grows, so too does the amount of data that stream into companies' Big Data collections. And some are already starting to feel the strain as they struggle to maintain performance levels across hybrid environments.

"IoT threatens to generate massive amounts of input data from sources that are globally distributed. Transferring the entirety of that data to a single location for processing will not be technically and economically viable," said Joe Skorupa, VP and Distinguished Analyst at Gartner. "The recent trend to centralize applications to reduce costs and increase security is incompatible with the IoT." (Gartner, Gartner Says the Internet of Things Will Transform the Data Center, March 18, 2014)

IoT Creates Complexity

Consider how a wearable fitness tracker works. The user interface is simple, but that little wristband is always working to send and receive information via Bluetooth from a smartphone, upload that information to a cloud-based app that analyzes activity levels, nutrition, sleep quality and heart rate, among other metrics. The application then supplies that analysis to its dedicated smartphone app, and likely also with other mobile and Web-based applications. And the user expects all of this to occur in real-time.

To meet these user expectations, network communication and interdependent application processes on distributed environments need to perform flawlessly. It takes only one piece of this application delivery chain to bring down the house of cards.

Think about a trucking company managing a fleet of delivery vehicles. It can install a variety of connected devices on a truck to monitor when it passes a checkpoint, its fuel and tire air pressure levels, how fast it is traveling, engine wear, and other information in real-time. This enables the company to provide real-time updates to customers, ensure the driver is driving safely, automatically schedule maintenance, and re-route a truck if bad weather looms. Now imagine how much more complex that operation becomes when trying to scale this across the entire fleet. So many things can go wrong within the application, infrastructure (cloud or on-premises), network or third party web services.

Recommendations

It is clear that IT needs real-time visibility into the performance of massively distributed applications, and implementing APM will enable you to:

1. See it: Monitor distributed applications and the underlying networks so you can quickly identify when the application, network or infrastructure impact end user experience. You will be able to examine the entire application topology and transactions that flow through it. from a transactional perspective: how is information being collected and shared between devices, applications, cloud services and the analytics systems.

2. Prioritize and Fix: Pinpoint the causes of bottlenecks or errors, and prioritize which ones you should address first. Analytics built into the APM product should be able to identify business-critical processes from the rest. Once you have done so, drill down into the offending code, network or infrastructure to gather diagnostics and resolve.

3. Be proactive: Because your APM product continuously monitors your applications, tracing all transactions, you have a wealth of information that you can analyze for patterns, to identify minor bugs before they become severe, or to seek opportunities to further improve performance.

As IoT devices feed more and more data to enterprises' Big Data stores, IDC predicts that within three years, 50 percent of IT networks will transition from having excess capacity to handle the additional IoT devices to being network constrained with nearly 10 percent of sites being overwhelmed. (International Data Corporation (IDC), IoT Market Forecast: Worldwide IoT Predictions for 2015, December 7, 2014)

Implementing an APM product will help you detect those constraints by identifying potential bottlenecks and ensuring business critical applications are constantly optimized and have adequate bandwidth. End-to-end visibility and diagnostics provided by APM is a necessity to identify issues with complex, distributes applications and to take action before users complain. Further, the detailed analytics it provides can help you be proactive about performance improvement, and to measure the business impact of the application.

Krishnan Badrinarayanan is Sr. Product Marketing Manager at Riverbed Technology.