[123776] in North American Network Operators' Group
Re: 10GBase-t switch
daemon@ATHENA.MIT.EDU (Joel Jaeggli)
Mon Mar 15 19:52:33 2010
Date: Mon, 15 Mar 2010 16:51:37 -0700
From: Joel Jaeggli <joelja@bogus.com>
To: George Bonser <gbonser@seven.com>
In-Reply-To: <5A6D953473350C4B9995546AFE9939EE08FE6635@RWC-EX1.corp.seven.com>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On 03/15/2010 04:30 PM, George Bonser wrote:
>
>
>
>> -----Original Message-----
>> From: Dave Temkin
>> Sent: Thursday, March 11, 2010 12:51 PM
>> To: Kevin Oberman
>> Cc: nanog@nanog.org
>> Subject: Re: 10GBase-t switch
>>
>> Can you point to another 1U box that has more than 16MB per-port
>> buffer?
>>
>> -Dave
>
> Anyone know what the buffer depth is on the Brocade TurboIron 24x? One
> thing to keep in mind is that neither the Arista nor the Brocade
> TurboIron are store and forward, they are both "cut through" switches
> which means they use less buffering for a given amount of traffic.
> Basically, a packet is already going out the egress interface before it
> is even completely received on the ingress interface.
Strickly speaking when mixed 1Gb/10Gb configurations are supported (as
the original poster was requesting) cut-through forwarding no-longer works.
> I have them (TI24X) in production and they work ok. They are lacking
> some features of the larger switches, though. No MRP or stp-group
> stuff. But they are a decent switch for the price.
>
> George
>
>
>
