How an F500 Health Services Firm is Using NeoLoad to Champion Performance during Enterprise Transformation
As the firm continues to expand its Continuous Delivery adoption practices across the organization thereby changing the pace and quality of its software delivery, performance has become the “often and early” measurement. Though ad-hoc load testing remains a high demand task at various times and for new projects, the Enterprise Performance Engineering team has developed creative techniques to rapidly test APIs early, enabling development teams to visualize longitudinal end-to-end performance data through the NeoLoad performance platform.
The entity we’re talking about is a leading information and technology-enabled health services business who dedicates themselves to helping make the health system work better for everyone. With >100,000 global employees, the firm’s delivers intelligent, integrated solutions which, they feel, help contribute to the health and performance of its customers.
In every digital transformation, change is a new norm. Agile, DevOps, and Lean methodologies optimize for small batches, short feedback loops, and a high degree of automation in software pipelines, requiring efficiencies from all layers in the product value chain. People, process, and technology all must evolve to stay in sync with changing customer demands. Delivering reliable, useful software into the Health Services industry is particularly challenging as customers are increasingly mindful of privacy and timeliness of technologies.
Security, high availability, and service reliability are increasingly crucial to the company’s consumers and partners. Zero-day patch performance, the seamlessness of scalability events, and delivery of defect-free service changes are real reasons for development teams to care about the performance of what they deliver genuinely. Responding to change quickly is less challenging when you have the right fit of skills, practices, and tools.
Why is Responding to Change so Important?
As the firm consistently improves its processes, changes to infrastructure, security, and code all present risk to the customer/workforce experiences. A recent new security initiative was introduced across all vital enterprise services and apps requiring the Enterprise Performance Engineering team to deliver comparisons and new baselines on performance to ensure a smooth and successful rollout.
Corporate security regulation compliance required that the performance testing team also create separate timing windows during which all testing can/not be executed. What did this mean for an already highly disciplined group of testers? Scheduling and resource coordination had to accommodate a 7 p.m. to 6 a.m. (CST) window. Delivering this quickly became an additional success metric. If time compression wasn’t hard enough, the testing days demanding higher load volumes grew infeasible with their legacy solution, LoadRunner. Unable to fulfill the new testing requirement resulted in the need to seek an alternative testing provider.
What is the Performance Impact on Security?
Security patches and improvements must be completed quickly and tested comprehensively. Updates can’t sustain the wait that legacy performance testing solutions require, and without visibility over possible service degradation, rollouts can be risky and costly.
Throughout three weeks during the Spring of 2017 with a NeoLoad testing portfolio of >30 web and mobile applications, ~20 APIs, the dedicated performance team completed its migration thereby meeting the delivery SLA. They also realized what they defined as “wins” using crucial NeoLoad capabilities such as mobile, Cloud/AWS testing, and NeoLoad Web.
(Example: Test results from the continuous performance testing of critical services.)
The team’s ability to respond to changes quickly and comprehensively, in part, lies in their decisions over which technologies they once utilized. According to the company’s Director of IT, Performance Engineering, “… NeoLoad’s plug-and-play UI was so easy-to-use ourselves that we were able to train colleagues within a few days quickly.”
So, in less than the two-week budget allotment, full API and end-to-end performance testing were completed using NeoLoad with ease. The team was able to roll out the new security measures with no user impact!
API Test Acceleration
The Enterprise Performance Engineering team has expanded its NeoLoad coverage to include to parallelize API functional regression testing. Since APIs typically present far fewer complexity barriers, the team constructs API tests in NeoLoad (for REST/SOAP), leveraging its high-concurrency engine to complete API regression tests in minutes, not hours.
As an outcome of this time-cost reduction, the firm’s development teams regularly obtain real-time streams of timely feedback on the performance characteristics of their changes (meeting the “early and often” objective). The team is now spending far less time scripting with NeoLoad; they have more time to support development’s appetite for continuous performance improvement.
(Longitudinal trending of their API Performance Tests via NeoLoad Web)
The team has also moved away from any notions of being “performance gatekeepers,” and instead, have become NeoLoad evangelists and trainers. The best part, the wider staff can spin up their performance measurement infrastructure including integrated dashboards in NeoLoad, deployed as Docker images, all combined with Jenkins and other CI orchestration platforms.
What Does it Take to Move to a Modern Performance Platform?
Beyond delivering high-performance systems, product teams need platforms which enable flexibility, shared knowledge space, and operational autonomy – moving away from backlogged technical debt due to a lack of timely performance visibility to a proactive approach that includes quick visibility has made collaboration easy.
The team also boasts of regular performance clinics as a necessary tactic that drives development teams to improve code efficiency which informs the infrastructure procurement planning processes. “To be able to help internal peers self-serve has represented a huge organizational win for us,” says the Director of IT, Performance Engineering.
Ask any of the team members who all share the sentiment that NeoLoad’s flexible, easy-to-learn, collaboration-friendly platform address three core company requirements:
- The enterprise-ready licensing model encourages team autonomy within guidelines
- Geographically distributed test and development teams can work through VPNs
- Product and vendor support for modern web, mobile, API and legacy technologies
Additionally, integrating stable performance test suites into continuously automated pipelines as more product teams employ this strategy ensures that performance feedback loops are fast, reliable, and maintainable directly within development cycles.
In an organization with as many associates at it has, it’s clear how NeoLoad, with its focus on integration ease and fast, and intuitive training curriculum has delivered the results required. Going forward, the Enterprise Performance Engineering team is excited to move beyond load testing and taking advantage of NeoLoad’s integrations with Dynatrace and Selenium to provide a robust, seamless end-to-end performance view. With this level of comprehensive performance data, developers can visualize in real-time the impacts of code and infrastructure changes. Life is good for this Health Services company!