Application Performance Management in 2013
By Andreas Grabner
Earlier this year, we at Compuware APM declared 2012 to be “the year of APM.” End users were demanding higher-performance applications than ever before, including faster page loads, e-commerce transactions, credit card authentications and smoother video streaming. The impact of poor application performance on the bottom line was a bundantly clear. APM was emerging as a top priority for CIOs, and businesses were getting the message that true visibility into the end-user experience was an essential component of an effective APM strategy.
With this sea change well underway, forward-thinking organizations are evolving their APM strategies yet again. They are using APM to improve application performance and identify bottlenecks and optimize resource utilization for supporting infrastructures. This helps drive business value. Organizations are also rounding out their APM approaches to include what were previously blind spots in the application delivery chain, such as the mainframe.
2013 Prediction #1: APM will find a niche in big data
The goal of big data is to find insights and make better decisions using data from traditional databases as well as from the fast-growing new sources of digital data, including the Web, biological and industrial sensors, video, e-mail, and social network communications.
Extracting timely business insights depends in large part on the availability and speed of the big data environment. Big data’s volume, variety, and velocity can greatly increase the overall complexity of the application and infrastructure which in turn can impact performance. The reality is that big data applications and environments suffer from many of the performance challenges and bottlenecks that plague current distributed applications, though the volume of performance data can be much larger. This puts ROI from big data projects at risk.
Poorly performing big data applications and environments can impact the business and revenues when customers complain or business analytics are delayed or unavailable. For example, is the issue with a particular Hadoop MapReduce job, data access or distribution, bad code, a server or hardware, network, or the application itself? Trying to scale away the problem by adding more nodes, clusters, or hardware can often be extensive and futile.
Increasingly, we believe APM will be used to help ensure the performance and availability of big data applications to meet business demands and satisfy end users. In addition, APM will offer deep insight into systems to optimize computing resources and data distribution across nodes, assure job execution efficiency, identify I/O bottlenecks, and tune CPU and memory consumption among thousands of nodes.
2013 Prediction #2: APM will increasingly “light up” the mainframe
The mainframe has long been established as a stable, reliable, and consistent platform for high-value, business-critical transactions such as enterprise resource planning, online order-taking and financial transaction processing. In recent years we have seen the mainframe’s role evolve, largely as a result of the “self-service” explosion on the Web.
For example, it used to be that mainframe applications did not directly interact with customers. Consider the banking industry. Banking customers used to walk into a bank branch to request their account balance, which a teller would then convey to them using a mainframe-based application to access customer data and process the transaction.
Nowadays, end users check their accounts directly and conduct basic transactions via the Web (and increasingly the mobile Web). To complete these types of transactions, mainframe-based applications are still used to access customer data and process transactions, but they are a part of a much larger and more complex set of systems known as the application delivery chain. Requests from millions of end users are hitting the mainframe every day, which impacts mainframe resources while simultaneously demanding exceptionally high levels of mainframe speed and availability.
Given the sheer number of mainframe transactions, subtle mainframe optimizations can exert an enormously positive and measurable impact in terms of optimizing expensive mainframe resources <em>and</em> ensuring superior end-user experiences. Organizations will increasingly realize they are remiss if they do not include the mainframe as part of their APM efforts.
Prediction #3: APM will be used to identify faulty third-party services, on the back-end as well as the front-end
Over the past few years, the drive to bring applications to market faster and more cost-efficiently has resulted in increased adoption of third-party services. Organizations often choose to leverage existing functionality from external third parties rather than develop it on their own. Some of these services -- such as social media plug-ins, product tours, and ratings and reviews -- sit on the front-end and serve to create a richer, more satisfying end-user experience. To date, organizations have made good progress in using APM to ensure high-performance for third-party services influencing their customers’ experiences.
Other third-party services like credit card verification provide critical functionality and reach all the way back to the back-end, actually entering the data center and the mainframe to complete transactions. The modularization of software has many benefits for organizations, but the problem with plugging in someone else’s code is the fact that it might not be optimized for one’s own environment, specifically the mainframe. Along with including the mainframe in their APM initiatives, we believe that in 2013 IT teams will work more closely with third-party service providers "reaching back" to the mainframe to ensure these services are tuned to their mainframe environments.
In 2013, we at Compuware predict that a next level of APM will truly emerge. It will focus on delivering exceptional application experiences while keeping costs in line. It will help organizations increase ROI and time-to-value for major initiatives like big data. Ultimately, the next level of APM will help bolster the bottom line, solidifying its place on the CIO’s priority list.
Andreas Grabner is senior performance architect and technology strategist at Compuware’s application performance management (APM) business unit.