Case studies

Achieving Scalable and Stable Performance: How Codvo Helped a Leading Payments Platform Build and Execute a Performance Testing Strategy

About the client

A leading fintech company based in the United States, providing point-of-sale payment solutions for retailers and consumers, partnered with Codvo on performance testing of its payments application.

The company's core business operations involve processing payments via APIs and rendering the user interface for consumers and retailers. Implementing a performance testing strategy for the application in-house was challenging, and the company worked with us for domain and technology expertise within the FinTech industry.

Overview

The company's core business operations involve processing payments via APIs and rendering the userinterface for consumers and retailers. Implementing a performance testingstrategy for the application in-house was challenging, and the company worked with us for domain and technology expertise within the FinTech industry.

Business Challenge

The client wanted to stabilize its highly functional application for retailers and consumers for better utility and improved performance. Their platform faced several challenges related to API responses and UI rendering, which were critical to their business operations. One of the key challenges was the limited system resources and access restrictions that hindered testing their application in a production-like environment. 

 

Due to the lack of performance testing, it was difficult to identify bottlenecks and optimize application performance. Performance testing and tuning would help with the platform's limits testingand issue resolution. The client was also looking for consultation on performance testing tool selection.

Our Approach and Solution

Our team first approached the problem by understanding its complexity and using technology to select the perfect tool for the client's needs within budget. 

 

To load-test the client's application, we decided to 

  • Use Apache JMeter and Locust
  • Two popular load-testing tools that simulate thousands of users  accessing an application simultaneously
  • Use Blazemeter to assist with API functional testing. 

To reduce the time spent on maintaining or creating test scripts, we created a separate Performance Testing framework for the API and UI, following a modular approach. To put the system's limitations to the test, we ran performance testing in a scaled-down environment and then extrapolated the results to match the production server's performance. 

 

We also used Jenkins and Gatling to automate the process of load testing and continuously monitor the application's performance. 

 

Application Performance Monitoring (APM) tool called Dynatrace allowed us to take a deep dive into each layer of the software stack to find the root cause of any performance issues. 

Tech Stack

The tech stack used by Codvo.ai for this project included: Java, Jenkins, Gatling, JavaScript, Scala, Dynatrace, BlazeMeter, Apache JMeter, Grafana, InfluxDB, Python, Locust, NeoLoad.

Highlights

Business Impact

Our approach helped achieve a scalable application with stable performance and minimal performance bugs for the client.

We could identify and resolve bottlenecks in the client's application leading to improved user experience and higher customer satisfaction.
Modular approach to the Performance Testing framework reduced the effort required to maintain or create test scripts, saving time and resources.
The client can now confidently release application updates, knowing they have been thoroughly tested and optimized for performance.