Keeping up with Apps: Network monitoring shifts to application performance focused on end-user experience

In today’s app-saturated business environment, the success of any business is largely determined by its ability to live up to great end-user expectations: optimum performance of business critical applications, and minimum downtime.

While the market has, over the years, seen a steady growth – both in number and sophistication – of tools for network troubleshooting, they have become insufficient. The intricacies of today’s increasingly complex network environments and multiple new applications being added every day on a variety of devices, combined with control protocols running on the backend that make those applications work, have made network monitoring more challenging than ever before. For instance, the behaviour of applications change when new versions are rolled out, making them more difficult to identify.

“Visibility into application performance is critical for organisations today, but it is getting more challenging to achieve it,” said Amit Rao, senior director, Business Development at Fluke Networks. “Today’s complex hybrid IT environments and applications span traditional data centres, third-party hosting facilities, branch offices, mobile devices, and SaaS and IaaS-based clouds. The evolution of networks to a hybrid model that combines private and public links further complicates the delivery of applications.”

The rise of NPMD

Network performance monitoring and diagnostics (NPMD) has recently emerged as a unique market designed to address these new requirements. Gartner, in its 2015magic quadrant for Network Performance Monitoring and Diagnostics, estimates the NPMD market at US$1.1 billion under the larger network management space ($1.9 billion in 2013), and overlaps slightly with aspects of the application performance monitoring (APM) space ($2.4 billion in 2013).

Unlike their traditional reactive counterparts, NPMD solutions have built-in application intelligence, which provides broad visibility and diagnostics that allow network administrators to pre-empt application issues through proactive monitoring, and identify root causes so they can act quickly to solve problems as, or before they arise. NPMD solutions also make capacity planning a breeze.

Highlighting the growing importance of NPMD as an emerging technology segment, Gartner produced the first ever Magic Quadrant on Network Performance Monitoring and Diagnostics (NPMD) in 2014. The first report emphasized the need for a single integrated solution to monitor, analyse and troubleshoot networks—and even more important—the applications and services they carry.

The recently published 2015 magic quadrant underscores the need for network professionals to understand the impact of network behaviour on application and infrastructure performance and gain insight into the quality of the end-user experience. This not only allows them to identify, troubleshoot and repair outages and degradations that affect user experience, but also optimise network and application performance.

Criteria in the quadrant, which identify the leaders and players in the NPMD space, include real-time performance, historical performance and predictive behavior capabilities, as well as flow-based data and network packet analysis. Vendors were assessed based on their ability to handle multiple application and technology types, offer optimum flexibility, and use SaaS delivery methods within portfolios. The latter gives enterprise IT teams more choices, while making previously premium-priced NPMD solutions more cost-effective for small and midsize businesses. Their solutions also offer a high degree of application-aware insight and visibility.

An example is Visual TruView, which has landed Fluke Networks in Gartner’s leaders’ quadrant in both 2014 and 2015 reports. This unified NPMD solution, which Fluke Networks also refers to as Application-aware Network Performance Management (AANPM), incorporates five tools in one, including real-time user experience monitoring, packet stream-to-disk, NetFlow/IPFIX collection, device health, and VoIP QoS. It also allows network professionals to gather data from multiple sources, providing both a wide breadth of visibility and the detailed forensic data required to solve challenging network problems.

To avoid confusing NPMD/AANPM with APM (application performance management), “PMD/AANPM takes a user-centric view of everything happening across the network, providing end-to-end visibility of both the network and applications and their interdependencies across Levels 1-7. It looks at how applications are deployed and how they’re performing, rather than simply looking at it from a coding perspective,” clarified Rao. “In contrast, Application Performance Management (APM) solutions usually support the application development team, providing auto-discovery of applications and transaction analysis, application usage analysis, end-user experience analysis, user-defined transaction profiling, and the basic functions of monitoring the health and performance of all configured application infrastructure assets.”

How NPMD/AANPM works

First, NPMD/AANPM provides network administrators a clear understanding of how well applications are traversing the network. This includes monitoring actual end-user transactions and measuring the speed by which the network, server and application, deliver the data to the end user. This application awareness comes from capturing the packets on the wire and monitoring their Quality of Experience (QoE) in real time.

The next level of visibility is understanding other important details like how quickly the transactions occur, who the users were, what function of the application they were using, what they were trying to accomplish, where the delays occur, and the like – before they start troubleshooting performance issues. This level of understanding, which gives them a clearer picture, involves the use of Deep Packet Inspection, or DPI.

Response time data is then tied into workflows that leverage traditional NPM data to look at flow, rate, utilization, endpoint details or even raw packet captures along the path of the user or application.

The road ahead   

The exponential growth in the use of mobile devices will converge wired and wireless networks, and in turn require solutions that can manage end-user performance expectations across both environments. Likewise, with unified communications solutions adding video calling capabilities into business communication tools, demand for bandwidth will continue to increase, requiring management solutions that can maintain performance of other business critical applications running across the same networks.

“The increased use of cloud and hybrid cloud models as methods of application delivery is lowering cost of acquisition, hence we’ll see more affordable NPMD/AANPM solutions in the future,” said Rao. “At Fluke Networks, we are on track to introduce network performance management solutions via the cloud.”

This NWA special feature is commissioned by Fluke Networks.