Re: [HSSG] Topics for Consideration: Jumbo Frames
Michael Johas Teener wrote:
> Jumbo frames are great as long as you stay in the HS environment. A jumbo
> frame hitting a 100baseT (or, heaven help us, a 10baseT) link will mess with
> any QoS assumptions.
That was the point of my question regarding the context of the
discussion on jumbos. This is also probably why Joel wants it in the
MAC objectives here. I don't have a problem with jumbos as long as
vendors ship the equipment with a default MTU of 1500B and leave
configuring the use of jumbos to those who need them.
> Nasty things will happen bridging to non-Ethernet 802
> LANs as well. I think this question needs to be addressed at a combo
> 802.1/802.3 level since it is a real architecture issue ...
> On 8/9/06 12:09 PM, "Joel Goergen" <joel@xxxxxxxxxxxxxxxxxxx> wrote:
>> My preference is to associate this with MAC objectives. In past meetings
>> where this has been addressed, the atmosphere was 'tense'. Thus best to get
>> the debate over with asap and move on.
>> I don't know at this point how to address your question. I suspect that since
>> 'all' of us support jumbos and we are inter-operable, that there is a
>> compelling argument to spec it but not change it.
>> Both my position and that of the customers I deal with is that higher speeds,
>> 10gbps included, is more efficient with the larger packets.
>> Mike Bennett wrote:
>>> Wouldn't jumbos be something better addressed separately? How do we deal
>>> with backwards compatibility?
>>> Joel Goergen wrote:
>>>> If I go to any search engine and input "jumbo ethernet frames chips", I will
>>>> see that every system and silicon vendor supports jumbo ethernet frames
Michael J. Bennett
Sr. Network Engineer
LBLnet Services Group
Lawrence Berkeley Laboratory