Put simply Appynamics tests the speed of an app, where bottlenecks are, and helps developers to fix issues – all in real time. This “dynamic baseline” — it uses machine learning that understands the apps load times and acceptable response times — can help speed it up.
AppDynamics’ vice-president of market development and insights, Jonah Kowall, was over from the US to brief analysts and media on why users are becoming impatient with response times and to reassure us that they had “absolutely nothing to do with #censusgate – although if they did then, IBM would not spell MUD. The remainder of the briefing is paraphrased.
Several years ago AppDynamics would have been an answer to a question that did not exist. Legacy hardware and monolithic software, low Internet bandwidth, and lower expectations of responsiveness were the norm.
Back then, 80% of IT budget was to “keep the lights on”. Today it has dropped to 45%. The priority now is the massive increase in spend on applications to 55% – call it digital transformation. In other words, IT’s driver is no longer so much about support as innovation. According to Forrester Research, “Firms are investing hundreds of millions of dollars in the wholesale transformation … to achieve a new level of customer-centricity and agility.”
IT, for want of a better descriptor, has become all about velocity. Amazon deploys a new “app” every 11.6 seconds, Facebook twice a day and banks maybe twice a year. Amazon and others in the know say that the old “big bang” has been replaced with thousands of smaller steps – velocity counts.
Forrester also says that 57% of users will abandon an app if it takes more than three seconds to respond and app performance drives customer loyalty and business performance. The app has become the business.
This has led to a new issue called micro-services, where hundreds to thousands of apps or APIs (application program interfaces like location, translate, etc.) are tied together to deliver a customer experience. Most Web pages call multiple — from a few to hundreds — of APIs to deliver the service. If even one of these APIs is unresponsive, it will make the app breach the new three-second rule.
AppDynamics call this an increasingly fragmented environment where traditional silos (Marketing, fulfilment, accounts, etc.) are replaced with micro-silos that can cause just as many issues.
AppDynamics’ approach to monitoring micro-services-driven applications, in an environment that shares space with conventional applications “monoliths”, involves what a “business transaction” - a chain of events that represent a service taking place — a request for data from a database, a response to that request that may contain the data, an order, fulfilment etc.
It says that whoever monitors services does not need to understand the architecture of the application, or contact the person who does. In monitoring the normal behaviour of the application, its App iQ can detect when and where API requests take place and ascertain and present key performance indicators for transactions inside and outside of micro-services.
Which is a long way to say that AppDynamics auto-discovers and maps the data flows, infrastructure, interrelationships and the environment to find the problems – unusually in less than 60 seconds.
AppDynamics is doing well. Revenue was up 86% to US$143.8 million. Its traditional competitors like BMC, CA, Dell, HP, IBM and Oracle have all had negative growth YoY.
To quote one client – the Nasdaq: “It was simple to deploy. It took minutes. On our first proof-of-concept, we saw results within an hour.”
iTWire interviewed Andrew Brockfield last year.