Latest Mountkirk Games Case Study (2024) - Google Cloud (GCP) Architect

To aid you in your study and test preparation, we will share our analysis for GCP Professional Cloud Architect - Mountkirk Games. The main focus of this case study is on Gaming analytics and Cloud Auto-scaling.

This case study has been around for a while and was recently modified on May 1, 2021 to coincide with the latest version of the Architect certificate exam. Let's discuss understand and analyze this case study

Company overview

Maker of online, session-based, multiplayer games for mobile devices. After successfully transferring their on-premises infrastructures to Google Cloud, they began extending to new platforms. A recent endeavour is to create a retro-style first-person shooter (FPS) game that allows hundreds of simultaneous players from many platforms and locations to join a geospecific digital arena. A global leader board of all the best players in each current arena will be displayed in real time on a digital banner.

you can refer the case study document at : https://services.google.com/fh/files/blogs/master_case_study_mountkirk_games.pdf

Key statement here are :

• Session-based multiplayer game builder for mobile devices. They began extending to other platforms after successfully migrating their on-premises systems to Google Cloud.
• A current idea involves creating a retro-style first-person shooter (FPS) game that will allow hundreds of players to simultaneously join a geospecific digital arena from various devices and locations.
• On a real-time digital banner, a global leader board including all of the greatest players from every active arena will be displayed.

Business requirements:

• Cut costs as much as possible.
• Provide support for a variety of gaming platforms.
• Provide support for multiple regions
• Keep delay to a minimum.
• Allow for quick game feature iteration.
• Take use of managed services and shared resources.
• Make adjustments to allow for dynamic scaling.

Technical requirements:

• Scale dynamically dependent on game activity
• Display scoring information on a global leaderboard in near real time.
• Save game activity logs in structured files for future reference.
• For multi-platform support, use GPU processing to render graphics server-side.
• Assist with the eventual migration of legacy games to this new platform.

Possible Architecture Solution

For building a Gaming Analytics cloud Platform solution we can consider below major components here :

• Kubernetes + Agones (Gaming Server) : For multiplayer gaming
• Cloud Storage : Store game activity logs in organised files for later analysis in the cloud.
• Batch : Cloud Storage > Dataflow > BigQuery
• Streaming : Pub/Sub > Dataflow > BigQuery
• Common infra : Cloud build, KMS, IAM, Logger etc

Reference Architecture Solution diagram

There will be multiple possbile solutions for this use case, to understand the implementation better we have give reference architcture which is one of possible solution
Below reference architecture depicts different aspect of Gaming Analytics cloud Platform solution like gaming server on GKE, Batch and stream pipleine and common cross cutting infrastructure components :