Mobile DevOps implementation that transformed the application lifecycle for the most loved game worldwide, Ludo King by Gametion

Client Profile

Founded in 2008, Gametion is one of the fastest growing companies in the Indian Gaming Industry. It is known for creating games that reach out to a variety of different audiences; something for everyone if you will. Their games are intended to be playable by various different age groups with each finding something to their liking. Ludo King being the most successful digital board game, ranking as India’s #1 gaming app on Android globally and No.1 game on App Store India in 2020.

Key Requirement

Gametion has published multiple games for iOS & Android under their umbrella. Ludo King is one of their most successful game with the multi-player feature. Gametion started experiencing issues with managing the increase in user traffic. The major challenge was that scaling wasn’t thought about when they first started and hence, they could not cope with the increased demand.
Here are the high-level challenges faced by Gametion team:

• No central code repository for mobile and backend application artefacts
• Multiple versions of code per developer
• Manual build and deploy
• Lack of streamlined release process

Flentas Approach

Flentas helped define and implement the overall DevOps processes, effectively and effeciently, thereby, addressing the challenges of continuous application delivery. We helped them with easy and smooth transitioning to AWS. We deep dived into the existing architecture and had to completely re-design which took us three months to implement. There was a need to release backend changes more frequently for small fixes or new features. Due to manual releases, it was difficult to manage deployments across a fleet of backend servers. Scaling at speed was made possible using the below AWS Services:

• AWS Code Commit was setup as a central code repository in cloud for versioning of code
• AWS Code Build was implemented for automating the dependency building of backend libraries
• AWS Code Deploy was setup with rolling deployment strategy for EC2 instances
• AWS Code pipeline was used to orchestrate the entire release pipeline for every code push on master branch
• Separate pipelines were created which triggered auto build and deployments for different environments (Dev, QA and Production) respective to GIT branches (Git Flow)

Outcome of the Project

Success Metrics

• Developers, Testers and Operations teams collaborate more effectively and transparently by being a part of the same automated process
• Increased frequency of releases which accelerated product improvement at a faster pace
• Automatic release and deployments reduces human errors and increases reliability
• AWS DevOps services have a pay-as-you-go pricing model and are server-less which makes them less burdensome from cost and maintenance perspective.
• Reduced user drop-offs by 99.99%