[BBLISA] Faster than 1G Ether ... ESX to ZFS

[Sean Lutner]

On Nov 16, 2010, at 7:51 AM, Edward Ned Harvey wrote:

>> From: Sean Lutner [mailto:slutner at rentul.net]
>> Sent: Monday, November 15, 2010 11:20 PM
>> 
>> ESX absolutely supports bonding (teaming). It's a standard best practice
>> configuration for your vSwitches if you want to do failover or outbound
> load
>> balancing. See the esxcfg-vswitch command for details. If you can and want
>> to setup port channels (in IOS land) you can also control inbound traffic.
>> There are plentiful blogs and vmware community posts about how to set this
>> all up.
> 
> This is true.  In fact yesterday while I was doing something unrelated, I
> stumbled upon those controls.  So it's pretty obvious, if you just give it a
> try.
> 
> However, the fact remains, that bonding/teaming will still be limited to one
> wirespeed for one connection.  So bonding/teaming might be useful if you
> have a bunch of  machines talking to a bunch of machines...  But it won't be
> useful if your ESX host is talking directly to one other host, for the sake
> of storage.

I run a somewhat sizable virtual environment with almost 50 ESX hosts and up to 50 guests per host. The environment has over 1,000 VMs and we don't have a single ESX host with more than 1GbE NICs. Our storage devices (Netapp) all have 10GbE out the back. We have never seen a bottleneck on the host side using 1GbE connections. We split the hosts into three networks; 1 for vkernel and storage (all NFS), 1 for guest networking and 1 for the service console/vmotion. These are all bonded/teamed failover pairs. seperating your traffic like this is also a best practice and something I can highly recommend you do. Having a full 1GbE connection dedicated to all these things is almost certainly more than enough. What are you doing in your environment that you need 10GbE?

> 
> _______________________________________________
> bblisa mailing list
> bblisa at bblisa.org
> http://www.bblisa.org/mailman/listinfo/bblisa
>