asked on

Mapped Drive vs UNC Performance Differences?

Is there any performance benefit to using Mapped Drives vs UNC paths?

I'm looking to see if there is less network or resource overhead to using a mapped drive in Windows vs a UNC path that may re-authenticate between calls.

Any documentation to back this up?

megs28

I'm not overly sure as to what you mean when you say UNC vs. mapped drives. I take it you mean typing '\\server\share' from a run prompt vs. mapping a drive. With that in mind, I would think that the mapped drive would cause more traffic due to the fact Windows will "check" to see if the connection still exists when you even do simple things such as browse My Computer. This in turn would cause more load for the OS as well (some people seriously wonder why it takes 2 minutes to open My Computer when they have 20 mapped drives that are rarely accessed). This amount of traffic shouldn't cause a noticable slow down, and if it is, you need to sit down and examine your infrastructure very closley.

By the way....UNC = \\server\share. You can map a drive by that, or substitute the server's IP address instead. And if this is what concerns you and slows your network down...again, you need to examine your infrastructure.

usacadena

using the static ip address should be the fastest \\10.1.#.#

brandonkirsch

ASKER

I know what a UNC is. I get what you're saying about opening "My Computer" and all but I'm looking for concrete facts (that is a link to documentation) to base our decision on.

We're making thousands of queries across SMB shares so every millisecond counts.

Changing to an IP address for \\10.10.* wouldn't make a difference since DNS lookups are cached anyways after the first call.

Does anyone have a link to documentation on the overhead involved with using UNC paths vs Mapping a drive letter?

ASKER CERTIFIED SOLUTION

usacadena

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial