Application performance monitoring (APM) and network performance monitoring (NPM) are becoming increasingly important as businesses that have adopt cloud-based services and virtualized infrastructure.
In the recent SDxCentral report, “Network Performance Management Takes On Applications,” more than half of surveyed respondents are actively looking at APM and NPM systems, and more than one-third are in the testing and deployment phases of adoption. Another 16 to 20 percent are piloting these systems, and roughly 15 percent have already deployed them in their network.
Among those surveyed, 37 percent were technology vendors, 24 percent were telecommunications service providers, and 19 percent were enterprise end-users.
The issue is the increasing complexity of networks as applications move off of traditional hardware and into the cloud.
The challenge is not only in providing visibility, but also in making sure the infrastructure “is being correctly configured with these apps lying on top of it,” says Andreas Grabner, performance and developer advocate with Dynatrace, an APM and NPM company.
It was much easier to have visibility before, when everything was being run on your own server on-premises. Now applications are being run on public and private clouds, which don’t provide visibility in all the areas that you need, Grabner explained.
Traditional network management tools only run discovery-based analytics periodically, creating significant visibility gaps in software-defined networking (SDN) and network functions virtualization (NFV) environments, according to SDxCentral’s report. Being able to pinpoint issues is becoming increasingly difficult since applications now run on a dynamic, automated infrastructure.
Tools Present Some Challenges
High-volume applications such as video streaming and the Internet of Things (IoT) are driving the need for scalable tools that can collect and analyze high volumes of application and network traffic, according to the report.
The majority of those surveyed in the SDxCentral report say that current APM and NPM tools are not meeting their needs. Many traditional APM tools only focus on custom-built applications and are therefore highly specialized.
“Many visibility companies want to have a wide range of tools to offer but are usually extremely limited,” says Andy Huckridge director of service provider solutions with Gigamon. “Customers may ask for very specific functions, which their tool vendor might not have.”
It is beneficial to be able to source tools from multiple vendors in order to avoid being locked in with one vendor who may not be able to deliver specific functions, Huckridge says.
IT teams need performance management that can provide end-to-end visibility into all application types running across hybrid environments. Performance management also needs to be able to find all types of application issues no matter where they are located, the report says.
Customers are also looking for a flexible delivery model that allows them to deploy tools in any combination of on-premises, software-as-a-service (SaaS), or hybrid implementations that can analyze a variety of performance data sources.
Network monitoring is needed for enterprises because they are typically offering services. Being able to make sure the network is running better than their competitors’ gives them a competitive advantage, Huckridge says.
“Monitoring is how you ensure that you are spending IT investments in the right place,” says Patrick Hubbard, head geek at SolarWinds. “If you are a larger organization, it is likely that you are spending a ton of money on your network, and they [monitoring systems] allow you to see if you are being cost effective or not and how you can be.”
According to the SDxCentral report, application performance can have a large impact on businesses. If a company’s application involves transactions or other revenue-producing functions, then it is crucial that it stays up and running at the highest possible level.
Grabner says Dynatrace is working to ensure that applications have the right quality and performance already built in. He adds that having these application-specific tools in the development stage and production stage gives companies a timely advantage.
“I think that monitoring is paramount in the industry, because otherwise you are flying blind,” Grabner says.