May 16, 2019 // By Magenic
Leveraging Magenic’s Automated Quickstart (MAQS), a DevOps-enabled automation deployment pipeline was implemented with a project-level testing process that incorporated scalable agile processes to increase efficiency and speed.
Challenge
A global audit, consulting, & financial advisory firm sought to expand, improve, and enhance an existing internal application that was becoming a popular and valuable tool for the organization.
The existing application was hosted in and managed by the organization’s data center, causing long development cycles and only two deployments per year.
The organization lacked a proper testing infrastructure or data management strategy. Testing was performed manually using spreadsheets at the end of a development cycle slowing down software release frequency.
Solution
Magenic utilized a risk-based approach to properly focus automation efforts. After an automation framework was established, the team was able to integrate automated testing with a new CI:CD process. This provided early feedback on build and deployment quality.
To reduce the load on the organization’s data center and facilitate more agility, the application was migrated to Azure.
The data management challenge was addressed by creating test utilities to perform data accuracy testing and harvest historical data for future business analysis. These tools allowed the team to input large amounts of randomized data, and to store results for future comparability to review accuracy and repeatability over time.
Results
Smoke Testing (37 web service and UI tests)
- Post-merge smoke testing reduced from 4 hours of manual testing to ~10 minutes using a suite of 37 test cases. Result was a standardized deployment pipeline that was less expensive and easier to maintain.
- Enabled more frequent deployments in the organization.
Regression Testing (71 web services, 130 UI)
- With a suite of 200 automated UI and web services test cases, regression cycle times reduced from 40+ hours to under an hour.
Custom Test Automation Utilities
- Data specific testing time reduced from 30+ hours of manual testing to under 6 hours. Accomplished with only 14 test cases running against several hundred data samples.
- Metrics produced during data testing were used to enhance the application’s ability to automatically identify data and patterns.
By leveraging the MAQS automation framework, initial framework set-up and implementation time reduced by ~30%.
MAQS at a Glance
MAQS is an open source framework created by the QA Engineers at Magenic that automates a set of QA tasks. It enables organizations to continuously deliver digital products faster, and with fewer defects, saving both time and money.
