Applications are growing more complex and living on a wider range of platforms than ever before, and the application performance-management space is evolving along with them.

APM platforms are providing more unified, flexible and predictive monitoring hubs for developers and operations, spurred to deliver more targeted application information faster. At the heart of this changing dynamic in APM is the rise in business stakeholders taking an interest in how a given application’s performance affects value.

“APM has been around for a long time, and the people who need software and performance analytics continue to expand by leaps and bounds,” said Abner Germanow, senior director of solutions marketing at New Relic. “We’re moving from software to automate existing business processes to software that’s actually the business.”

In a recent survey entitled “DevOps and the Cost of Downtime: Fortune 1000 Best Practice Metrics Quantified,” IDC analysts identified the rising influence of DevOps in application performance and found that application downtime costs Fortune 1000 companies between US$1.25 billion and $2.5 billion every year. Conducted in cooperation with AppDynamics, the survey breaks down the lost value of application failures down to the hour, while projecting that the total number application deployments will double within two years.

Kalyan Ramanathan, vice president of product marketing at AppDynamics, comes from the mobile and DevOps worlds. Formerly of mobile APM provider Crittercism and DevOps automation company Electric Cloud, Ramanathan talked about the IDC survey in the context of how customer experience has become the most important deliverable and driver for DevOps in terms of business transactions—any user action within an application that has a business outcome.

“APM solutions are at the heart of measuring and managing customer experience,” Ramanathan said. “It’s fundamentally about understanding the business impact of performance because software applications today are not just an IT entity; they support the business. You have to instrument all the components of the application, gather data from them all and tie the business transaction and components together with complete visibility into the performance.”

Modern APM trends drive a changing philosophy
As DevOps practices and more prevalent business concerns change the way organizations view application performance management, the concept of what APM is supposed to be has also changed. What was once a market synonymous with back-end applications now encompasses mobile, Web and cloud applications, and often more than one rolled into the same application as different components. Aruna Ravichandran, vice president of DevOps marketing at CA Technologies, sees mobility as a crucial factor in any modern APM platform.

“More and more of our customers are launching mobility initiatives to monitor performance in mobile apps,” she said. “With APM solutions like mobile app analytics, you want to provide the end-to-end view. It’s not enough to be able to monitor the performance of Web-based applications; you need to manage applications native to mobile devices, connected to some kind of back-end application, which is then connected to some kind of mainframe. A big element of our strategy is to integrate mobile app analytics with service virtualization to provide continuous feedback to the developer throughout the application life cycle.”

AppDynamics’ Ramanathan highlighted these underlying platform shifts as well, pointing out mobile applications, cloud-based deployment and server-side architectures as three emerging technology frontrunners that are making the prospect of managing application performance markedly more complex.

“Just about every application and CIO today is looking at cloud-based deployment, whether it’s internal, hybrid or something like AWS,” Ramanathan said. “If you’re an application vendor today without a mobile front end, you’re pretty much dead on arrival. We’re also seeing server-side architectures truly take off. Dependencies on APIs and micro services today are going through the roof. A typical app now has six to 10 APIs it talks to on the back end, from internal APIs to external services like Facebook, Twitter and PayPal. Apps are no longer these monolithic things; they’re composed of a lot of services, and APM platforms need to take an end-to-end view to understand them.”

Within the diversifying application landscape, deployments are growing ever more frequent. According to a New Relic survey, 58% of its customers deploy application code weekly or even faster. In that heightened atmosphere, APM solutions are finding simpler ways to present data, not only to developers but to all the stakeholders within an organization, and curating that targeted application data in real-time using both real and synthetic user monitoring.

“We surface the data we think is most relevant to what a developer or a business stakeholder wants,” said New Relic’s Germanow. “Rather than flooding you with data only an APM specialist would understand, we create a curated experience. You have people who want visibility into different aspects of your application, and it’s a need that goes beyond performance where everyone has a different set of questions.”

To go along with this faster, more specific need for real-time data in solving application problems, APM is also trending from a reactive philosophy to more-predictive tooling and notifications. According to CA’s Ravichandran, the market shifted from the days of being reactive in the early 2000s, to a point in the mid-2000s where the APM market has started trending in a way that’s much more predictive. APM platform focus has shifted toward forecasting in finding anomalies and correlating behavior to fix problems as soon as or before they occur.

“The traditional APM story is a reactive story,” said Buddy Brewer, vice president of business development at SOASTA. “Rather than root cause analysis, now we want to focus on the predictive side. If you have a problem, it’s important to react quickly and fix it. But what’s better than that is to avoid the problem in the first place because you found it before you went to production.”

In providing these various real and synthetic user-monitoring platforms along with predictive analytics, the goal of many APM vendors is to redefine application crisis mode by eliminating one of its most prevalent symbols: war rooms. This trend toward predictive solutions, informed by in-depth and tailored user data, is guiding APM from a reactive to a more proactive approach in rapidly addressing application problems.

“The real value today is not so much to fix problems after they happen, it’s to prevent problems from happening in the first place before you put an application into production,” said Nicolas Robbe, CMO of Dynatrace. “We don’t believe the goal is to have better war rooms; we want to have no war rooms. We want companies to turn their war room into a ping-pong room.”

Dev, Ops and business impacts
“There are many different personas now within the APM domain,” said CA’s Ravichandran. “You have an application support person, the system administrator, the service delivery person and many more. To truly make the promise of DevOps real, you need to facilitate better communication and collaboration between the Dev and Ops teams.”

Robbe explained that after reshaping around the relationship between developers and operations, a third dimension of APM is now coming into focus: business. Rather than deriving abstract intelligence about an application, platforms are tackling APM from the standpoint of user experience, and how any small change in application performance might affect a user’s business value.

“We’re now seeing the business side showing interest in the APM data from the standpoint of the user experience,” Robbe said. “Is a customer’s behavior affected by performance? If someone is complaining on Twitter about their experience, can I go back and figure out what happened? All this high-fidelity architecture is now providing this new layer of value to address another set of needs. We think of APM today as a digital performance platform, more than just an application-management platform.”

In that vein of a digital performance platform, AppDynamics’ Ramanathan talked about the notion of an application intelligence platform. These new age APM platforms combine unified monitoring tools affording developer and operations visibility into not only the code of an application, but also how end users are experiencing it, and how the user experiences factor into identifying and fixing performance issues in the business transactions flowing through the apps.

“DevOps alone is not the most complete way to approach APM,” said Ramanathan. “The notion of DevOps today is Dev and Ops working together to fix the right issues, the issues that are truly mission-critical. It’s important for Dev and Ops to look at problems by prioritizing the business considerations affecting the end user. The tactical value is narrowing down the thousands of issues you might spend oodles of hours trying to solve to the five issues that truly affect a customer.”

At the core of how developers and organizations approach APM are changing customer expectations. When DevOps teams serve business needs, and the business is concerned with how application performance affects users, the user has the power. SOASTA’s Brewer believes that in this context APM as a concept is more important than it’s ever been, it just means something slightly different.

“People are far enough away from remembering the dial-up era that they expect Web pages and mobile applications to be responsive,” he said. “It doesn’t matter how they’re connecting, whether they’re using a browser or a laptop, waiting in a line on their phone or using a tablet in their living room. The expectations get faster and faster in an environment where more and more revenue is coming from digital. Application Performance Management itself isn’t an outdated term. It’s that the components within APM and where people spend their time that has changed from being associated just with instrumenting the back end. There’s a lot more to that picture in 2015.”

Experts offer three developer recommendations for weighing APM platforms
Aruna Ravichandran, CA Technologies vice president of DevOps marketing:

  • Ease of use: “How easily can a developer get their hands around a tool that’s easy to deploy, easy to install and easy to manage? Having a simplified APM tool that developers can use to manage and troubleshoot is extremely important.”
  • Quickly pinpointing performance issues: “One of the biggest elements when you think about APM is complexity. As you add more features, the product becomes more difficult to use. We have a unique user-friendly dashboard that enables users to get access to agents across APM deployments on the developer side, and you can quickly generate diagnostic reports and diagnose any kind of problems, which is critical. We also use one-click notifications, so every time there’s an issue, a developer will receive underlying context of the performance issue explaining the key actions they need to take so they can diagnose and fix the problem quickly.”
  • Smart instrumentation: “Smart instrumentation is something a developer should completely leverage. It automatically connects transaction cases when problems occur, so you only collect the pertinent traces and don’t need to recreate issues. Smart APM instrumentation provides targeted snapshots that give actionable info. The benefit developers get is that with automated transaction tracing, you don’t need to have in-depth knowledge of the application or modify your instrumentation.”

Kalyan Ramanathan, AppDynamics vice president of product marketing:

  • Focus on the code: “What developers want first and foremost is to write amazing code, write lots of code and spend less time on maintenance and other distractors. So from an APM solution, they’re looking for a single pane of glass so they have great visibility into the performance of the app itself. They want to have a common understanding of the app’s performance to share with operations to make DevOps possible.”
  • Similar Dev and Ops tooling: “When you do performance testing of the app in a test environment, you’re using similar tools, concepts and definitions as would be applied to the production environment; a consistent view of how the application would work across both environments in the Dev and Ops parts of the organization. We have a solution called the Virtual War Room, a shared collaboration canvas that both Dev and Ops can use to look at the same issue and identify root causes more effectively.”
  • Deep-dive information: “When an application is moved to production and the team is working on managing the experience, when things break, developers need deep-dive information to identify the root cause of the problem. What Dev hates most is Ops coming to them and saying, ‘Your application does not work without providing any context of what really went wrong, where it went wrong and what they need to solve.’ ”

Abner Germanow, New Relic senior director of solutions marketing:

  • A holistic platform: “Which platform is going to allow developers to focus on their app more than the APM platform itself? It needs to be fast, secure and needs to be able to monitor the components regardless of where they run. You need to be able to look at a service not just in the individual APM realm, but an end-to-end view of real user performance in a way that gives insight into the back end, mobile, browser, and also synthetic use.”
  • RUM and SUM: You need a mix of both real user monitoring as well as synthetic monitoring, and you need to be able to get the data in a way that doesn’t require a lot of APM expertise. What happens when something breaks at 3 a.m. and no one is actually using the application? Can you get alerted to that before your users find out? It needs to enable the entire team to get immediate feedback loops on the parts of the application they care about.”
  • Web and mobile focus: “App architectures have changed with customer experience pushed to mobile and rich Web applications running on cloud services. Legacy APM and infrastructure-centric monitoring strategies were not built to monitor modern architectures on cloud services. Select APM tools that let you and your team focus on your application and customer experience from mobile and browser through the back end, not on maintaining an APM tool and monitoring infrastructure.”

Buddy Brewer, SOASTA vice president of business development:

  • Business context: “We’re past the era where you can have sets of data that don’t connect back into your business. If you’re working with part of the business, every set of data you collect connects back into your business somehow. When you’re selecting an APM tool, go with one that gives you some kind of business context. Whatever is happening in your application stack right now or over the last month, how did that impact your business and your revenue? You need that context to evaluate the trade-off between, say, working on performance this week and pushing off new features to next week.”
  • Real-time, for real: “Developers need tools that give them information in a timely manner. Real-time has meant different things to different people over the years. People used to get away with saying something was real-time within 15 minutes, but with flash sales, social media and how much brand damage can occur if you have an application performance problem, that’s not acceptable. When things happen in the real world, how long do you have to wait before your APM reporting lets you know?”
  • Predictive capabilities: “The reactive part of the APM problem is something all solutions more or less provide today if you choose any of the vendors out there. How can you use the APM information to learn things about your application that ultimately prevent problems from happening? Evaluating solutions along that predictive curve is important.”

Nicolas Robbe, Dynatrace CMO:

  • Data fidelity: “From a developer perspective, which is different from operations, you need to be able to understand how problems build up over time. Problems don’t happen all of a sudden. If your APM platform is operations-centric, it would only capture data when things go wrong. What led you to the problem is what you need to understand. Developers need the ability to collect every piece of data and retain it for as long as necessary to find and troubleshoot code instantly.”
  • DevOps tool-chain integration: “One of the ways you prevent performance problems from reaching production is by moving performance indicators up in the continuous delivery chain and creating gates based on reaching those indicators. So as you build a release, you would be able to know without having to deploy that application if it meets certain criteria that indicate an acceptable performance level.”
  • Take it for a test drive: “Go beyond the marketing literature and try it. There’s a tendency out there from some vendors to bypass that step, but the devil is in the details. Not all APMs are born equal, so trying out platforms and making sure the one you choose fits your needs and has the capabilities you need is crucial.”

A guide to APM offerings
AppDynamics
Application Intelligence Platform enables software-defined businesses to proactively monitor, manage and optimize complex application environments in real time, in production. The solution enables users to see everything with Unified Monitoring, from data center applications and infrastructure to browsers or mobile apps; act fast with DevOps collaboration; and know the business impact with Application Analytics with cloud, on-premise and hybrid deployment flexibility.

AppNeta provides Full Stack APM, delivering integrated performance visibility that spans the application code, through the network, to the end user. AppNeta’s SaaS solutions give Development, Application and IT Operations teams broad, detailed performance data to see across their Web, mobile and cloud-delivered application environments and pinpoint performance bottlenecks.

BMC Software delivers software solutions that help IT transform digital enterprises for competitive business advantage. For IT Operations, TrueSight App Visibility Manager goes beyond application performance monitoring to provide deep insight into user experience. In addition to tracking and measuring user activity at the individual or location level—on premises or off—it filters data to ensure the delivery of relevant application information without the unnecessary noise.

CA Technologies’ CA APM is built to be easy, proactive, intelligent and collaborative, or EPIC, offering easy deployment of APM agents, including PHP. Proactive identification and resolution of issues occur across physical, virtual and mobile applications. Intelligent insight comes through 360-degree mobile-to-mainframe visibility, which captures billions of critical metrics per day to verify transactions. CA APM is collaborative, aligned with DevOps methodologies to instill continuous performance improvements at every stage of the software life cycle.

Catchpoint Systems offers innovative, real-time analytics across its Synthetic Monitoring and Real User Measurement (RUM) tools. Both solutions work in tandem to give a clear assessment of performance, with Synthetic allowing testing from outside of data centers with expansive global node, and RUM allowing a clearer view of end user experiences.

Crittercism’s mobile application performance-management solution enables enterprises to accelerate their mobile business. Crittercism’s solution monitors every aspect of mobile app performance, and provides a real-time global view of crash, service and mobile transaction metrics across iOS, Android, Windows Phone 8, HTML5 and hybrid apps.

Dell Foglight APM delivers answers, not just data. Get up and running in minutes with a smart APM SaaS-based solution that requires zero configuration, or monitor your applications from a single pane of glass with an on-premises solution that can be optionally combined with other Foglight performance-monitoring capabilities for database, virtualization or storage to support a full range of enterprise monitoring strategies—whether your application is deployed in the data center, cloud or hybrid cloud.

Dynatrace delivers developers and testers the answers they need to ensure superior user experiences, proactively solve problems, and accelerate DevOps/Continuous Delivery strategies. Dynatrace User Experience Management and Dynatrace Synthetic Monitoring help developers to proactively understand and optimize user experience; Dynatrace Application Monitoring is designed to maximize application performance; and Dynatrace Data Center RUM delivers app-aware network insights.

HP’s Application Performance Management suite empowers businesses to deliver exceptional user experiences by examining the impact of anomalies on applications before they affect customers. Business applications are available with proactive end-user monitoring and actionable diagnostics to measure the performance and stability of apps from a user perspective.

Neotys’ NeoSense is a proactive application performance-monitoring solution for all Web and mobile applications. It actively monitors the performance and availability of critical business transactions within recorded user paths. NeoSense leverages the test scenario design capabilities of NeoLoad to quickly create realistic monitoring profiles. Easy-to-read dashboards, alert triggers, and notification rules provide actionable insights for pinpointing performance issues and getting to root causes quickly.

New Relic’s comprehensive SaaS-based New Relic Platform provides one powerful interface for Web and native mobile applications. Leveraging the New Relic APM, code-level visibility for applications in production that cross six languages—Java, .NET, Ruby, Python, PHP and Node.js—and supporting more than 60 frameworks, the platform also includes real-time with New Relic Insights along with mobile, browser and synthetic user monitoring capabilities.

Oracle provides a complete end-to-end application performance-management solution for custom and Oracle applications. Oracle Enterprise Manager is designed for both cloud and on-premises deployments; it isolates and diagnoses problems fast, and reduces downtime, providing end-to-end visibility through real user monitoring; log monitoring; synthetic transaction monitoring; business transaction management and business metrics; Java and database monitoring and diagnostics; multilayer discovery of infrastructure and application topology; and application performance analytics and advisories.

SmartBear Software’s AlertSite UXM monitors the connected world of Web, mobile, and SaaS apps and APIs for businesses from 80+ global locations. AlertSite provides fast, accurate alerts on problems before end users experience them, including issues with third parties. Monitoring scripts are easily created in minutes—no programming necessary—with just a few clicks of the mouse or by importing API test scripts from SoapUI by SmartBear.

SOASTA believes real-time user data gives companies the most actionable intelligence needed to stay in front of performance issues before they impact the business. The SOASTA platform enables digital business owners to gain continuous performance insights into their real-user experience on mobile and Web devices—in real time and at scale.