“We aim to find a common language”

Originally written in German by Martin Szelgrad for Report

Standardization of cloud infrastructure and self-healing IT: Linz-based software company Dynatrace researches and develops technologies for future software generations.

If Dynatrace were active in a division other than IT, many would know the Linz-based unicorn well, given its revenue of $487 million, most recently achieved in 2020. Having grown from a software startup to a global market leader in just a few years, the Austrian-American company with headquarters in Massachusetts and a research center in Linz is a leader in the field of application performance monitoring. Their products ensure the optimization of IT processes and are thus something like the micro brain for the digitization of companies in all industries.

“Imagine you are a retailer and during Cyber Monday your mobile app or website crashes. What would you do?”

as asked during the announcement of a partnership with SAP and its Commerce Cloud solution suite in October 2020. Thanks to Dynatrace solutions, retailers know very well what is happening in their IT environment — where problems are occurring in their shopping platform and how the performance of individual digital services affects the bottom line.

Alois Reitbauer is Chief Technology Strategist at Dynatrace and is responsible for areas and topics that will not be immediately transformed into new products. They do, however, chart the path of the technology company in the medium term. These include the company’s commitment to open-source software, various technology partnerships, and also its own research branch, which Reitbauer is currently setting up.

“One of the big topics in the software community is the monitoring and supervision of software systems,” Reitbauer explains. The technical term “observability” is used to monitor and document applications and IT infrastructures for their functioning and performance quality at all levels of IT processes — from the server in the data center to the user’s personal computer or mobile application. Dynatrace has made a name for itself with its ingenious products in precisely this area. The software company from Austria is world-renowned in the IT scene. Its toolbox for application performance monitoring has been constantly expanded and today also includes diagnostic solutions for cloud environments, including features that use artificial intelligence to make it possible to trace where and, above all, why errors occur.

In December, Dynatrace also entered the security market for cloud applications, helping companies find and fix blind spots. The future, Reitbauer says, lies in automation: systems that counteract and repair themselves when performance bottlenecks occur.

Self-healing systems

One of the groundbreaking projects on Reitbauer’s plate involves cloud infrastructures that can manage themselves. In a world of increasing networking, however, this requires standards that go beyond the product worlds of individual market participants. Dynatrace is a welcome partner to Microsoft, Google, and other household names in the “OpenTelemetry” project. “We want to find a common language for IT processes — starting with the way monitoring works in general to the so-called ‘tracing’ of processes in huge IT systems,” he describes.

Reitbauer has also been active in the international standardization body W3C for years. Its chairman, Tim Berners-Lee, is a co-founder of the World Wide Web. The Linz-based company is also the only Austrian company involved in the Cloud Native Computing Foundation (CNCF), which has produced technologies such as the Kubernetes container format. “In our Keptn project, we are now developing an automation layer for operational IT systems that can also handle immensely large amounts of data. We are bringing in experience with our own products. Because, with the thousands of systems worldwide in which our performance monitoring is plugged in, [operations] will no longer be possible without automation,” he explains. One of CNCF’s credos is to make the results of its work openly and freely available to the market — which, among other things, is what has made Kubernetes so successful.

Automation tools for IT infrastructure already exist. What these all lack, however, is uniform semantics and thus interoperability when moving between systems or cloud environments. In another international collaboration, Dynatrace is in contact with Google, Ericsson, and others with the goal of ensuring the “continuous delivery” of IT processes.

Research division

Dynatrace started thinking big years ago, looking outside of the company’s own boundaries. “We have reached a level of complexity with our product that requires us to develop our own basic technologies, similar to Microsoft or Google,” says Reitbauer. The Linz-based company has created a technology called “Davis AI” that uses explainable artificial intelligence to support the comprehensible analysis and filtering of large amounts of data across different structures. With increasing floods of data, many companies face the challenge of even recognizing what data they have. “If the business department doesn’t have an overview here, it won’t be able to ask the right questions about new business opportunities,” the technology strategist is convinced, and he also makes it clear that humans alone will no longer be able to keep an overview of it all.

Data sizes in the exabyte range (one billion terabytes) can no longer be handled satisfactorily with conventional database queries. “When I started at Dynatrace, projects with a hundred servers were the gold standard. Today, clusters of hundreds of thousands have become normality,” says the Austrian, who spends 50% of his time in Silicon Valley during non-Covid times, as he describes the almost unfathomable pace of growth in IT everywhere. With millions of servers as sources in the future and a wide variety of data formats, there is a need to react to changes in milliseconds or at least in periods of a few minutes. In these dimensions, conventional analysis tools reach their limits. It does not matter whether the purpose is to optimize infrastructures or to forecast the next sales success: the technology supports its users in mastering the challenges of everyday work.

With Davis AI, huge amounts of data are pre-filtered and presented to users in an appropriately reduced form. For example, in an “alert storm”, when performance problems arise in an IT application. “We will see more intelligent decision systems that automatically ensure the desired state of applications and restore them if necessary,” he is convinced. The operating instructions for this are already being written in machine-readable form today. Alois Reitbauer is now building up a research branch at Dynatrace, with an emphasis on active cooperation also with the local science and developer scene, such as at the Linz Institute of Technology (LIT) Open Innovation Center. Employees are intensively being recruited for this venture.

Dynatrace’s customers include Porsche Informatik and Walmart in the U.S., which use the Linz-based company’s automated tools to optimize and secure their services and business processes — whilst gaining insights from the flood of log data, requests and observations.

Be that as it may, the future of many companies lies in data. And an Austrian company is actively involved in this field.

“We aim to find a common language” was originally published in Dynatrace Engineering on Medium, where people are continuing the conversation by highlighting and responding to this story.