How much does the Relay service cost?

Note: There is no cost associated with using Relay during the open Beta.

Two factors determine the total cost of using the Relay service. The first is the average concurrent users (CCU) during a given month. The second is the amount of egress data from the Relay service.

Relay offers a free tier covering up to a set amount of concurrent users and data. Once this limit is reached, your project is moved to the paid tier and you will be billed for additional concurrent users and data. For more information, please see UGS pricing.

How Relay defines and captures the monthly average CCUs

Periodically, every Relay server snapshots the number of players for a given project. The snapshot of all servers is summed up together into a number of player/CCU seconds since the start of the month for a given project. 

For example, if 10 players were active from 10:00 AM to 10:00 PM (12 hours) there will be an average of five CCU total for that day. If this is repeated every day in a month, there will be a total of 5 monthly average CCU.

The following scenarios would all lead to a monthly average CCU of 50:

1. 100 players 24/7 for the first half of the month, none for the remaining half.
2. 50 players 24/7 for all days of the month.
3. 100 players for 12 hours per day on all days of the month.

How Relay defines and captures the monthly data consumed 

Relay bills by outgoing bytes only and the bytes egressing the Relay server. That means all traffic going to the host, and to every player, combined. 

For example, a host that sends a total of 50 KB to eight clients produces 50 KB of egress from the Relay server. Presumably, the host also receives X bytes from those eight players, which would be added on top of that egress. 

In this example of a three-player game where one player is also the host, we have 15 packets per second (PPS) sent between players. We can see the combined input/output that the host receives and sends out: