A fleet is a collection of servers that host a game or application in specific regions. After your fleet is online, you can start making allocations via your matchmaker. You can have one or more fleets, and each fleet has a view where you can view the fleet details, the linked build configurations, the scaling settings, and the fleet analytics.

Each fleet has the following associated information:

• Fleet name: The human-readable name of the fleet.

• Fleet ID: The unique identifier of the fleet.

• Operating system: The operating system that the machines within the fleet run. You select a fleet's operating system when you create a fleet. You can't change the fleet operating system later.

• Build configurations: The build configurations linked to the fleet.

• Regions: One or more regions and their associated scaling settings.

Fleet example

A studio is releasing a new multiplayer game, MP-B, which is hosted on Multiplay. This studio had previously released a game on Multiplay, called MP-A. Both games have two components that need hosting:

• The game server, which is where players actually play the game session.

• A training server, which is where players learn how the game works before joining a real session.

Each of these games must be billed separately, so Multiplay creates an account service for MP-A and another for MP-B. For both games, the training servers use fewer resources than the game servers, and don't require as much scaling. Accordingly, Multiplay separates these two servers into different fleets: one fleet for training servers, and the other for game servers. The following diagram shows an example of this configuration.

Additional resources

• Fleet lifecycle

• Fleet analytics

• Fleet API workflow

• Machine provisioning