Last week my sister and I stopped at a local restaurant chain, and an employee stepped out to greet us. “I’m sorry,” she said. “The system’s down, and we can’t serve any food today.” We stopped at another location, and the story was the same.

I immediately put on my IT hat and thought that the problem was centralized since it was affecting multiple locations. But on further circumspection, I surmised it could have been a broad-based network problem that was affecting the company’s major system of record (SOR). I was also imagining the revenue this chain was losing by the minute.

When the network is the data

Network-related problems will increasingly come to bear on big data. Network professionals prefer not to hear this, because they already have their hands full taking care of network service levels for SOR applications. But to the extent that companies increase their reliance on real-time streams of marketing and performance big data, the network will become a central part of big data application performance.

This is already having a major impact on IT and its service level guarantees. Organizationally, more CIOs are taking a serious look at reorganizing their departments so that a combined application performance management (APM) group that consists of applications and network professionals works together to support the app — whether it is SOR or big data.

The move is not a comfortable one for IT pros who are used to having their own silos of expertise, and are not used to working on a departmental team with employees who have different IT specialties.

In addition to the impact of reorganization on IT culture and workflows, CIOs also have to think about a new set of forces that can affect network performance and, ultimately, the ability to acquire and to act on real-time or near real-time big data.

Why this is important

A transit system in New Zealand monitors track and travel conditions of its trams in real-time and automatically sends out alerts to customers and to maintenance technicians when an impasse in the system is detected. Technicians get dispatched, and customers get advised over their mobile devices on alternate routes they can take to their destinations.

A large retailer measures consumer response to an online promotion in real-time and gauges which items are generating the most sales. The retailer has the option of immediately pushing out a new promotion for either the most popular items or less popular merchandise that will further generate sales. This is revenue capture that never would have been possible in the days of static promotions, where there was no way to respond to revenue stream activity while the stream was in process.

A pharmaceutical company ships environmentally sensitive drugs via air freight and continuously monitors sensors within the packages containing the drugs to ensure that temperature and humidity remain steady. If there is a breakage in the package seal, or if the environmentals begin to fail, the sensor immediately issues an alert, and the situation can be remedied before the valuable cargo spoils. A failure in any of these circumstances not only cuts off revenue, but can also mean the difference between life and death. And while sensors can be failed over and internal systems can be fail-proofed, what do you do if the wide area network (WAN) that flows over public internet and is beyond your control fails?

This is why incorporating network monitoring should be on your company’s big data road map if you anticipate using live streaming and analytics of big data and Internet of Things (IoT) data in mission-critical business applications.