The service quotas for endpoint groups in AWS Global Accelerator are as follows:
- The maximum number of endpoint groups per accelerator (across all listeners) is 42. This limit is fixed and cannot be increased. However, the total number of endpoint groups multiplied by the number of listeners must not exceed 42, which means the number of endpoint groups you can use decreases as you add more listeners[1][2][3].
- Each endpoint group must be associated with a single AWS Region, and all endpoints within that group must be in that same Region[4].
- The number of endpoints allowed per endpoint group depends on the type of endpoints:
- Endpoint groups with more than one endpoint type can have up to 10 endpoints.
- Endpoint groups containing only Application Load Balancers can have up to 10 ALBs.
- Endpoint groups containing only Network Load Balancers can have up to 10 NLBs; this quota can be increased upon request.
- Endpoint groups containing only Amazon EC2 instances can have up to 10 EC2 instances; this quota can be increased upon request.
- Endpoint groups containing only Elastic IP addresses can have up to 10 Elastic IPs; this quota can be increased upon request.
- Endpoint groups containing only Amazon VPC subnets can have up to 10 subnets; this quota can be increased upon request[1][2][3].
- Port overrides per endpoint group have a default quota of 10 and can be increased upon request[1][2].
- Other relevant quotas include listeners per accelerator (default 10, adjustable), standard accelerators per AWS account (default 20, adjustable), and custom routing accelerators per AWS account (default 10, adjustable)[1][2][3].
- Note that quota management and increase requests for Global Accelerator are handled in the US East (N. Virginia) region, while API operations must be run in the US West (Oregon) region[1][3].
In summary, endpoint groups in AWS Global Accelerator are limited to 42 per accelerator, with each group restricted to a maximum of 10 endpoints of a given type (with some quotas adjustable upon request). Endpoint groups must be region-specific, and the total number of endpoint groups and listeners combined must not exceed 42. Port overrides and other related quotas also apply and can be adjusted as needed through AWS Service Quotas.
This detailed quota information helps in planning and scaling your Global Accelerator deployments effectively.
Citations:
[1] https://docs.aws.amazon.com/global-accelerator/latest/dg/limits-global-accelerator.html
[2] https://docs.aws.amazon.com/general/latest/gr/global_accelerator.html
[3] https://docs.aws.amazon.com/en_us/global-accelerator/latest/dg/limits-global-accelerator.html
[4] https://docs.aws.amazon.com/global-accelerator/latest/dg/about-endpoint-groups.html
[5] https://docs.aws.amazon.com/general/latest/gr/servicequotas.html
[6] https://repost.aws/knowledge-center/globalaccelerator-limit-endpoint-access-by-securitygroup
[7] https://docs.aws.amazon.com/pdfs/global-accelerator/latest/dg/global-accelerator-guide.pdf
[8] https://docs.aws.amazon.com/global-accelerator/latest/dg/about-endpoints.html