The Black Hole in Digital Performance Management

by   |   February 22, 2017 5:30 am   |   1 Comments

Jason Bloomberg, President, Intellyx

Jason Bloomberg, President, Intellyx

For today’s enterprises, digital performance means far more than a snappy website and mobile apps. In reality, end-to-end performance is critical to keeping customers happy and, in turn, achieving the strategic goals of the business.

End-to-end digital performance presents a tricky challenge, as every link in the chain must operate properly. Finding weak links quickly and resolving them immediately can be the difference between an online purchase and a lost customer.

The Rise of Digital Performance Management

To gain the necessary visibility, operations managers turn to a variety of monitoring tools. Web performance management for the front end, application performance management for keeping track of numerous apps, network management to ensure the underlying pipes are free from clogs—the list goes on.

This piecemeal approach to operations monitoring, however, has proven insufficient for that ensuring end-to-end digital performance meets the increasingly stringent requirements of today’s users. The result is a burgeoning new market category of digital performance management (DPM).

The goal of DPM is to raise the operations monitoring discussion above the technical weeds and focus instead on the success of business transactions. Whether it is a purchase on an ecommerce site, a balance inquiry on a mobile banking app, or a physician viewing the chart of a patient on her tablet, business success means the performance of business transactions as they traverse all of the components of a modern, digital enterprise environment.

For today’s enterprises, however, “modern” doesn’t mean “all new.” Modern digital enterprise environments are a heterogeneous mix of old and new somehow wired together to support the rapidly changing digital landscape. That wiring, in fact, is what keeps the lights on. Without it, all the flashy digital doodads will amount to naught.

What is that all-important wiring that keeps the transactions at the heart of flowing? Middleware.

Middleware: The Black Hole

black holeOf all the odds and ends that make up today’s IT infrastructure, middleware is perhaps the least sexy but among the most mission critical. The motley assortment of technologies from over the years that fit in this category include message queuing products dating from as early as the 1980s, enterprise application integration technologies from the 1990s, enterprise service buses from the 2000s, and open source streaming-centric middleware like Apache Kafka and Spark from the 2010s—to name a few.

For operations personnel responsible for ensuring that digital business transactions perform properly, the mix of different vintages of middleware in the typical enterprise form a gaping black hole that most monitoring tools don’t illuminate.

Yet in spite of being hidden from view, middleware is absolutely essential to the organizations that depend on it for connecting applications, executing managed file transfers, and completing business transactions.

This middleware black hole is a problem that a global provider of financial services technology solutions, ran into. During one two-month period, the monitoring tools it had in place failed to catch bill payment anomalies—a serious business problem.

An internal investigation uncovered the problem: issues with older middleware its monitoring software had missed.

To solve this problem, Fiserv installed Nastel AutoPilot®, a monitoring tool that can capture middleware-centric issues that threaten the success of end-to-end business transactions. “Unlike our former system, AutoPilot enables our company to detect infrastructure problems that could impact our bill payment and fraud applications on our UNIX, Linux, and Windows platforms before they become full-scale problems,” according to Tony Konczak, Senior Mainframe Sustaining Engineer at Fiserv.

In fact, AutoPilot provided Fiserv with greater visibility into several middleware components and their metrics, including queue managers, command servers, repository managers, the dead letter queue, publish and subscribe brokers, and listeners. AutoPilot also provided application process monitoring of the queue depth, open input count, and open output count for Fiserv’s aging message queuing technology.

The technical benefits of such monitoring translated directly into business benefits for Fiserv. “The solution’s proactive engineering efficiency reduces errors and overall risk, gives us the ability to more quickly verify our clients’ threshold values, lowers support costs and, in general, significantly enhances overall quality,” Konczak added.

Filling the Hole

Business transactions are rarely simple, synchronous message exchanges between point-to-point application connections. Generally, they consist of complex, asynchronous communications.

Such long-running, multi-step asynchronous transactions typically cross the entire IT infrastructure, frequently traversing one or more corporate firewalls on the way.

Furthermore, various types of middleware brokers frequently route such messages to many destinations, including databases, servers, apps, application program interfaces (APIs) (both on-premises and in the cloud), and even mainframes.

Lightweight message tracking tools generally drop the ball at the broker, as it frequently transforms messages, making it difficult to keep track of them with rudimentary tagging techniques.

To solve this problem, AutoPilot examines and correlates method calls and message payloads to understand any pending or existing breaches in expected behavior and performance. The result: middleware is no longer a black hole, as operations personnel now have end-to-end visibility into mission critical business transactions.

Remember, Customers Depend Upon Your Middleware

Historically, the operations team responsible for the care and feeding of your middleware has rarely been in a position to directly impact your customer experience, and with it, your strategic business goals. With digital transformation, however, that long-standing reality has changed.

The central tenet of digital performance is end-to-end, from customer touchpoint to system of record. Every piece of software along this route must operate at top speed, every time—including the wiring that holds everything together, the middleware.

Don’t fall into the middleware black hole. Include it in your digital operations strategy instead; your customers will thank you.

 

Copyright © Intellyx LLC. Nastel is an Intellyx client. At the time of writing, none of the other organizations mentioned in this article are Intellyx clients. Intellyx retains full editorial control over the content of this paper. Image credit: Leonid Mamchenkov.

 

Jason Bloomberg is the leading industry analyst and globally recognized expert on agile digital transformation. He writes and speaks on how today’s disruptive enterprise technology trends support the digital professional’s business transformation goals.

He writes for Forbes, his biweekly newsletter the Cortex, and several contributed blogs. He also helps technology vendors and service providers communicate their digital transformation stories. His latest book is The Agile Architecture Revolution (Wiley, 2013).

Mr. Bloomberg has published over 750 articles, spoken at over 350 conferences, webinars, and other events, and has been quoted in the press over 1,500 times.

 

Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise.

Tags: , , , , , , , , ,

One Comment

  1. Luis Zea
    Posted February 22, 2017 at 11:16 am | Permalink

    Great article highlighting some frequently overlooked technology. Middleware’s important role in supporting complex chains of software that power today’s digital services cannot be ignored or taken for granted.

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>