Contoso intends to peer an East US VNet with a West Europe VNet. They are evaluating cross-region peering for an application with heavy inter-region traffic. Which statement accurately reflects bandwidth, latency, and cost considerations for global VNet peering?

The correct answer accurately describes global VNet peering. Azure uses its global backbone for cross-region peering, which offers high bandwidth but still incurs latency based on geographical distance. Data transfer between peered VNets across regions is billed as inter-region egress. Throughput is limited by the VM's network interface card (NIC) and SKU, not an arbitrary peering cap. The first incorrect option is wrong because global peering does incur inter-region egress charges, and performance is not always equivalent to same-region peering due to geographical distance. The third incorrect option is incorrect because there is no 100 Mbps bandwidth limit for global peering, and ExpressRoute is not required to scale beyond such a limit. The fourth incorrect option is incorrect because VNet peering does not automatically compress or accelerate traffic, nor is there an "accelerated peering" option.