Skip to main content

Amazon GameLift Streams

Amazon GameLift Streams features

Get started with Amazon GameLift Streams

Page topics

Game compatibility

Open all

Amazon GameLift Streams uses WebRTC open standard protocol which helps you stream your game without modifications. You only need to upload your application to an Amazon Simple Storage Service (S3) bucket, without the need to recompile or change your game code.

Amazon GameLift Streams supports games running on Microsoft Windows, Ubuntu Linux, or Proton with easy onboarding, providing compatibilities with game binaries at the right price-performance required for your game and business model.

Amazon GameLift Streams includes a custom launch script for Proton that enables you to bring your own pre-configured Proton game builds and helps you configure and test your game’s compatibility setting prior to deployment.

Save on cost

Open all

Amazon GameLift Streams offers three classes of compute resources—or stream classes— to choose from, giving you the flexibiliity to choose the right type for your game performance requirements and desired price point. A stream class represents the type of compute resources to run games. Stream classes vary in CPU, GPU, RAM, and other specifications, offering different levels of performance and cost. Amazon GameLift Streams provides NVIDIA GPU-based stream classes with 5 different performance options. The stream class you choose impacts streaming performance and cost.

For Linux and Proton runtimes, Amazon GameLift Stream classes offer multi-tenant stream classes, a cost-effective option that shares a single GPU across multiple concurrent streams. On Gen4 and Gen 5, the service offers one muti-tenant stream class “high”, and on Gen6 it offers three multi-tenant stream classes, “high”, “medium“ and ”small”. Multi-tenancy is useful for streaming games that don't require maximum hardware capabilities, helping you optimize your costs.

Scale and operate

Open all

Amazon GameLift Streams enables you to deploy and stream games across multiple applications and geographic locations from a single stream group in your primary region. Developers can share this capacity with multiple games using the same runtime (Windows, Linux, or Proton). Amazon GameLift Streams is available in the following AWS Regions: US East (N. Virginia), US East (Ohio), US West (Oregon), Asia Pacific (Tokyo), Europe (Frankfurt), and Europe (Ireland).

Multi-application stream groups allow you to link multiple game titles, enabling you to stream different games from the same pool of compute resources. This many-to-one relationship allows you to efficiently manage resources across multiple applications or versions of your game. When starting a stream session, you can specify which linked application to stream, supporting faster iterations and shared capacity between multiple game versions or different content offerings.

Track your streaming operations with detailed CloudWatch metrics. Access comprehensive data about stream usage, session durations, location distribution, and more. These insights help you optimize your service usage and measure game success.

Monitor individual stream sessions with real-time performance stats. Access detailed data on CPU, memory, GPU, and VRAM usage for an active session through the GameLift Streams WebSDK, or view them in the AWS console's built-in overlay on the "Test stream" page. Performance stats can also be exported to a file for post-session analysis. These session-level insights help you optimize performance and troubleshoot individual user experiences.

Developer tools

Open all

The Web SDK contains all the front end and integration pieces you need to get up and running with Amazon GameLift Streams. Included is the Amazon GameLift Streams example web application which is a sample HTML5 frontend local website that you can use to share your streams and accelerate your testing of Amazon GameLift Streams.

Monitor your game's performance and troubleshoot issues by collecting output logs from your game engine. These logs help you identify crashes and resolve production problems, providing valuable diagnostic information when needed.

When faced with complex technical issues, initiate a snapshot of an active stream to gather detailed resource data. This feature tracks filesystem changes from the start of the session to the end, and re-exports any modified or newly created files to your specified S3 destination. This allows you to examine the total system state and capture any binary artifacts which might have been created by your game’s built-in debugging tools.