A single-Region deployment hosts an app on Amazon EC2 that must accept both HTTP and non-HTTP TCP connections. The company wants to use the AWS global network to reduce latency and also place the instances in an Auto Scaling group behind an Elastic Load Balancer. What architecture should a SysOps administrator implement to satisfy these requirements?

The correct option is to launch an Auto Scaling group behind a Network Load Balancer (NLB) and use AWS Global Accelerator, registering the NLB as an endpoint. An NLB is required because the application needs to accept both HTTP and non-HTTP TCP connections, which an ALB cannot handle for non-HTTP traffic. AWS Global Accelerator improves performance by routing user traffic through the AWS global network, reducing latency. Registering the NLB as an endpoint for Global Accelerator allows it to leverage this optimized routing. The other options are incorrect because: An Application Load Balancer (ALB) only supports HTTP/HTTPS traffic, not general TCP. CloudFront is a Content Delivery Network (CDN) primarily for caching web content and does not natively support non-HTTP TCP connections for origin traffic. While it can front an ALB for HTTP/HTTPS, it's not suitable for the non-HTTP TCP requirement.