Windows Server 2012 NIC Teaming Part 3 – Switch Connection Modes

Windows Server 2012 NIC Teaming Part 1 – Back To Basics

Windows Server 2012 NIC Teaming Part 2 – What’s What?

There are 4 possible basic configurations of a NIC team, depending on your physical switches and how you want to distribute or load balance traffic across the team members of the NIC team.  In this post, I want to focus on the switch connection modes.

This configuration of the switch connection mode is determined in one of two ways:

  1. You decide how you want traffic to flow, that determines your switch architecture, and you design the team appropriately.
  2. You already have a physical switch architecture, and you have to configure the team appropriately.

There are two switch connection modes in a WS2012 NIC team.  My tip: focus on the use of independent and dependent when trying to remember which is which.

Switch Independent

This type of NIC team has no dependency on functionality in the connected physical switch(es) to make the NIC team work.  It is appropriate to use this type of NIC team in two scenarios:

1) A single dumb switch, like the sort you might get in a store for a lab

image

The switch does switching and that’s it.  There’s no management port or console, and no settings you can configure. The team works independently of any non-existing functionality in the switch.

Alternatively, the team members are plugged into multiple independent switches.  In this case, the switches might or might not have some clever management.  The key piece here is that each access switch is functioning completely independent of the other – there is no switch stacking going on.

image

A nice feature of switch independent teams is that you can configure a hot-standby team member in the NIC team.  This is only possible in a switch independent team.

Switch Dependent

The name says it all; the NIC team relies on some functionality in the switch(es) that the NIC team members are connected to.  This could be a single managed switch.  It could also be a single logical switch, such as a switch stack.

image

There are two ways to set up a switch dependent NIC team.  Both options require you to configure the switch(es) in some way (consult your network documentation):

  1. Static teaming
  2. LACP

Static teaming is when the switch ports are configured to be in the same team.  Using the above example, the switch ports for pNIC1 and pNIC2 would have to be configured to be in the same team.  This is pretty inflexible: try reconfiguring the team or moving the cables to different switch ports and you’ll break the team without doing some switch reconfiguration to match the changes.

Link Aggregation Control Protocol (LACP) is similar to static teaming because it requires switch configuration. However, once the switch is enabled for LACP, the team dynamically configures the switch whenever it comes online or is reconfigured.  This means that you do not need to constantly log calls for the network admins when you are doing physical server operations.

That’s enough for today.  Next up will be load distribution.

This information has been brought to you by Windows Server 2012 Hyper-V Installation and Configuration Guide (available on pre-order on Amazon) where you’ll find lots of PowerShell like in this script:

image

 

Windows Server 2012 NIC Teaming Part 4 – Load Distribution

Windows Server 2012 NIC Teaming Part 5 – Configuration Matrix

Technorati Tags: Windows Server 2012,Hyper-V,Networking

7 thoughts on “Windows Server 2012 NIC Teaming Part 3 – Switch Connection Modes”

  1. What are the benefits of LACP over independent? I’ve got an independent team that’s working great, but if LACP provides better bandwidth, perhaps I should try to get that going.

  2. Hi Aidan,
    You deliver excellent blogs and very useful articles.
    About NIC teaming, is NIC LBFO implementing IP probing ?
    I created a 2 NICs switch independent team. Each NIC is connected to a switch. Each swicth is connected to to a core swicth.
    Suppose that the active team member is the one connected to first swicth. What if the cable between this switch and the core swicth is unplugged (or the core swicth fails). The server will loose connectivity to the LAN, but the NIC will will not fail (no hardware or link failure) and the team will continue sending traffic to this NIC.
    The question is : Is there any method to workaround this (like IP probing in Intel for example)?
    Thanks Aidan

  3. Aidan,
    When I add my NIC Team to a Server 2012 external Hyper-V virtual switch, the team immediately stops sending packets. The receive packet statistics continue to increment, but the send statistics are reset to Zero and stays at zero. Any thoughts?

    Appreciate any assistance that you may be able to provide!
    Mike

  4. hi
    i am having a problem with network
    i have 4 nics
    2 are realtek gbe and 1 is intel 82575eb(dual port)
    i have created a team of 1 relatek and 2 ports of intel 82575eb many times and reformatted server many times, obtained drivers from hardware manufacturer and also installed all updates. tried windows 7 and 2012 r2.
    if single realtek nic, connection speed is 1gbps and all the diskless clients boot very fast but if use team of 1 realtek and 2 intel 82575eb or only 2ports of intel 82575eb it really sucks.1 client boots in 4 minutes while just single realtek 1gbps it takes to boot all 25 units around 1 minute and 30 seconds or less.
    please help me out
    waiting for your kind response.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.