DATEV eG is one of Europe’s largest software and cloud services companies specializing in tax management solutions. About 2.8 million businesses, freelancers, municipalities, and associations across Germany rely on DATEV’s products and services.

Summary

Challenges

Aimed to deliver reliable cloud-based services.
Wanted to speed the adoption of performance testing capabilities.
Sought to replace fragmented, custom tools with a centralized testing solution.

Solution

Enabled comprehensive and easy-to-use performance testing.
Empowered developers and testers with a standardized self-service solution.
Provided high levels of flexibility.

Results

Completed 17 times more tests every week
Detected 4x slowdown in a critical internal service
Improved the developer experience and drove adoption

Challenges

Aimed to maintain high levels of performance for cloud-based products
Wanted to ensure customer-facing services continued to function properly under high loads
Sought to enhance the quality, speed, and security of software development

DATEV eG is a large software and cloud services company headquartered in Nuremberg, Germany. With 8,900 employees, the company specializes in solutions to streamline and simplify tax processes in Germany and is cooperatively owned by 40,000 tax advisors, auditors, and law firms. As one of Germany’s and Europe’s largest business software providers, DATEV generates annual revenues of €1.44 billion and serves 2.8 million businesses, freelancers, municipalities, and associations.

Building on its many years of success in delivering desktop applications, DATEV now aims to offer its solutions via the cloud. The company is continuously expanding its cloud portfolio, regularly launching new apps and features to strengthen its market-leading position and grow into new markets.

Peer Witthöft, test engineer at DATEV, said, “As we deployed more apps into the cloud, we looked to give our developers quick and easy ways to bulletproof the performance of our cloud applications. To help ensure high customer satisfaction, we want to ensure a fast and consistent user experience even under high load.”

Building on agile best practices for functional testing and automation, DATEV aimed to make performance testing more accessible to its developers and provide them with actionable insights. Helmut Lukasczyk, quality engineer at DATEV, said, “Continuous integration and continuous development [CI/CD] is a key part of our software development process. To further improve software quality, speed, and security, we looked for a complete, easy-to-use performance testing solution that would help to bring quality-related topics to the fore.”

Our teams are using analytics from OpenText Core Performance Engineering to track performance and improve product quality. It’s much easier for them to store and access key metrics and reports from scenario runs compared to the custom tools some of them were using before.

Helmut Lukasczyk
Quality Engineer, DATEV eG

Solution

So developers can build high-quality cloud solutions and boost the company-wide adoption of performance testing, DATEV uses OpenText Core Performance Engineering to deliver powerful performance testing features with a streamlined developer experience.

Products deployed

OpenText™ Core Performance Engineering

Ensure application quality with scalable, cloud-based testing

Service deployed

OpenText™ Professional Services

Get a trusted partner to guide your information management path

Enabling comprehensive and easy-to-use performance testing

With OpenText Core Performance Engineering, DATEV can make it easier for its developers and testers to performance test large numbers of REST APIs. “When we evaluated the options, we quickly understood that OpenText Core Performance Engineering is the best solution for us,” said Witthöft. “The OpenText solution offers a comprehensive set of features out of the box, including monitoring and analytics. What’s more, it’s very easy to use.”

The OpenText solution builds on modern technologies such as JavaScript, which developers at DATEV use heavily when developing realistic performance testing scenarios. The company leverages the DevWeb protocol to build DevWeb scripts to reliably and efficiently write, share, and run advanced load tests. Because JavaScript is an established language at DATEV—and with a wide range of helpers’ tools and libraries available to increase developer productivity—the OpenText solution is easy to work with for many DATEV development teams.

Empowering developers and testers with a standardized self-service solution

A core testing team at DATEV works with product teams across the business to help them get the most from performance testing—for example, by providing guidance on industry best practices. With its self-service capabilities, OpenText Core Performance Engineering enabled the core team to standardize and onboard new product developers and application testers quickly.

These capabilities include the Virtual User Generator (VuGen) tool. Using VuGen, DATEV developers can refine their performance test configurations independently, creating test scripts that faithfully emulate user behavior.

Providing high levels of flexibility

Some internal API endpoints at DATEV cannot be reached from outside of the company’s highly secure data center. To solve this challenge, OpenText Core Performance Engineering enables DATEV to leverage specialized on-premises load generators alongside highly automated and scalable cloud load generators to test otherwise inaccessible API endpoints. Cloud load generators makes it easier to run multiple tests in parallel to speed up execution and gain new performance insights even faster. Dynamic, automated provisioning of cloud-native resources eliminates the need to manually set up and maintain various components—such as controller machines—for every test.

The flexibility of OpenText Core Performance Engineering allows DATEV to address a wide range of performance testing requirements. “The architecture of OpenText Core Performance Engineering gives us the freedom to test all kinds of APIs and user scenarios,” said Lukasczyk. “For some of our APIs, we use on-premises load generators. Thanks to the way these load generators are integrated into the solution, we can run them on a Kubernetes cluster in our high-security data center without increasing our external attack surface.”

A person monitoring hologram data in computer

When we evaluated the options, we quickly understood that OpenText Core Performance Engineering is the best solution for us. The OpenText solution offers a comprehensive set of features out of the box, including monitoring and analytics. What’s more, it’s very easy to use.

Peer Witthöft
Test Engineer, DATEV eG

Results

With OpenText Core Performance Engineering, DATEV is encouraging its developers to make performance testing a continuous process, promoting greater quality, performance, and resiliency for its cloud solutions.

Completed 17 times more tests every week

Since moving to OpenText Core Performance Engineering, DATEV has inspired more of its developers to make regular testing part of their standard process.

“Switching to OpenText Core Performance Engineering has been a big success across our cloud service teams,” said Witthöft. “Today, about 40 different cloud products with approximately 400 API endpoints in total are being tested continuously. Thanks to the improved usability, the cleaner workflows, and the high degree of flexibility of the OpenText solution, we now see at least 50 test runs from different teams every week—17 times more than before.”

Software quality has always been important at DATEV, and advanced functional testing as part of CI processes has long been routine across all products. With the move to OpenText Core Performance Engineering, the company aims to achieve the same frequency and coverage for its performance testing.

“More and more cloud product teams use performance testing as part of their CI pipelines,” said Lukasczyk. “Offering a comprehensive, easy-to-use standard solution, we improved DevSecOps practices by making performance testing a more integral part of the development process. It’s gratifying for us to see how the mindset of product development teams changes at scale, helping us to future-proof our cloud services and ensure reliability and availability even under high loads.”

Detected 4x slowdown in a critical internal service

While rolling out the performance testing solution to more teams, DATEV realized that it could also use OpenText Core Performance Engineering to improve internal infrastructure, such as its mission-critical authentication service. Lukasczyk said, “When we noticed slow response times for our authentication services, we decided to have a closer look. Thanks to advanced monitoring and the root cause analysis features built into OpenText Core Performance Engineering, we quickly identified the bottleneck that was causing a four-fold slowdown.”

Based on this success, DATEV plans to expand performance testing to cover more of its internal systems. “While product quality is crucial, ensuring fast performance of basic infrastructure services improves the experience for all our staff and improves the resiliency of the entire business,” confirmed Witthöft.

Improved the developer experience and drove adoption

Today, many development teams across DATEV have replaced fragmented, custom-built tools with easy-to-use, self-service capabilities from OpenText Core Performance Engineering.

“The self-service approach empowers developers to explore and try things out on their own,” said Lukasczyk. “For example, our teams are using analytics from OpenText Core Performance Engineering to track performance and improve product quality. It’s much easier for them to store and access key metrics and reports from scenario runs compared to the custom tools some of them were using before.”

Looking to the future, DATEV plans to adopt cloud load generators to enable more realistic performance testing scenarios while reducing management workload.

“Maintaining our on-premises load generators costs time and money,” concluded Witthöft. “Moving to cloud-based load generators will reduce our overhead while also offering access patterns over public networks that are closer to real users—delivering more realistic insights into potential performance challenges and bottlenecks.”