The challenges were to build a system oriented on:
- High level of Service Availability – no major downtimes expected for vital project
- Reduced Time-to-Market – more frequent delivery of new features and updates to keep pace with and even be faster then the rapidly changing e-commerce landscape
- Operations Excellence – not only faster Time-to-Market but also high quality of the final product – bug testing by customers is not allowed!
Dubai based fast growing B2B wholesale marketplace with predictive analytics. Company works for UAE, EU and Caucasus markets and plans to bring the product to new markets in Asia. The daily turnover of goods on the marketplace is more than $5M.
To address the existing and future challenges whole approach to development and operations processes were changed:
- Everything as a code – using Terraform and Github Actions to fully automate Project Builds and Delivery
- Ever green master – due to creation of aligned dev/stage/prod environments and implementation of E2E automated testing
- Using managed services with high availability where it’s possible – all project workloads were migrated to AWS and monitoring was set up with Datadog
- Balance the efforts of engineering team – SLI/SLO Policies were created & followed as reliable indicator to ensure the right tech points are always addressed but never overengineered
Customer got an overall growth of infrastructure health indicators for the last 6 months, as well as more efficient use of the engineering team:
- Availability of services increased to 99,3%
- Number of sev0 incidents dropped to zero
- Engineering Team Velocity increased by 25%
- Higher Product Quality – number of incidents in the ticketing system decreased two and a half times
- Costs & Time for Manual QA execution reduced by 70%