MicrosoftAs you have correctly pointed out in your question, VPN traffic speeds may be higher or lower than Internet traffic. Depending on network configuration, your assertion that VPN is faster may be true.
I certainly pointed out no such thing. I was very careful to stay away from this sore spot. I restricted myself to two scenarios: Internet to end-user, and Internet to WSUS.
What I did not detail in the article is the extensive work Microsoft has done on the infrastructure supporting Internet delivery of updates to clients (PC’s). Over the past 5 years, Microsoft has rearchitected our Windows Update Services and the Servicing stack within the OS to be able to deliver the minimum package needed for the device/machine.
And I am sure this "extensive work" represents millions (or even billions) of dollars of expenditure in both implementation and monthly maintenance. (If I was one of the Microsoft's members of the board, I'd ask: Why? What justifies this extravagant expenditure?) Still, it does not justify our migration from the Internet-to-WSUS scenario, to the Internet-to-end-user scenario. This migration multiplies that bandwidth cost by the factor of one hundred (the number of devices). What benefit do I get that justifies this extravagance?
