At its core, analytics for games is similar to analytics for other applications in different industries. You measure and track data elements to monitor performance, understand customer satisfaction, monitor sales, and so on. What differs is that analytics may not be a core compentency for your team depending on the size and maturity of your studio. Games are generating more data than ever, and you might not have a data analytics expert on staff to implement an analytics pipeline and manage infrastructure. You might not have dedicated expertise with skills to perform complex data processing and analysis. Another challenge is that many off the shelf solutiosn can create data silos and lack flexibility, making it hard to centralize data.
To address these common industry challenges, the Game Analytics Pipeline AWS Solution was created by Games SAs at AWS. This solution helps game developers launch a scalable serverless data pipeline to ingest, store, and analyze telemetry data generated from games and services. The solution supports streaming ingestion of data, allowing users to gain insights from their games and other applications within minutes. The solution is a CloudFormation template that deploys the following architecture:
This solution is entirely serverless. Serverless pipelines don’t require you to provision, scale, or manage any servers. You can use serverless for nearly any type of application or backend service, and everything required to run and scale your application with high availability is handled for you. A lot of companies are making the move towards serverless, even for their analytics workloads.
AWS has an expansive ecosystem of services that can be used to build out a robust analytics pipeline for your game so that you can discover useful insights on player and game data. Serverless is a great option for a lot of games companies because many of them do not have dedicated resources or a data analytics expert on hand to configure infrastructure and manage a data analytics pipeline.
The entire pipeline is modeled as “Infrastructure as Code” using a CloudFormation template. This automates the setup and provisioning of all the necessary services for the solution. You can use it as-is or you can download the source code from GitHub to extend and customize the solution.
The solution deploys AWS Glue to provide a data catalog which provides structure to the ingested data. You can access and analyze this data using the tools of your choice. The solution starts you off with Amazon Athena and Amazon QuickSight, but you can use whatever 3rd party analytics tools you want by using the Glue data calatog as a drop-in replacement for your external Apache Hive Metastore.
This workshop will focus on building out a simplified analytics pipeline similar to the Game Analytics Pipeline architecture above. This is to provide hands-on experience with the services used for analytics on AWS. Continue on to the next section to understand what architecture will be built and the services that will be used.
To learn more about how to deploy and use the Game Analytics Pipeline Solution, check out the Deployment Guide