Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [HSSG] Topics for Consideration: Jumbo Frames

We get upwards of a 60% performance gain when using jumbos for encapsulating fiberchannel onto ethernet.  Clustering and storage are our main performance gains when using jumbos. 







From: Mike Bennett [mailto:mjbennett@xxxxxxx]
Sent: Wednesday, August 09, 2006 1:25 PM
To: STDS-802-3-HSSG@xxxxxxxxxxxxxxxxx
Subject: Re: [HSSG] Topics for Consideration: Jumbo Frames


Joel Goergen 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.

that sounds reasonable. I just don't want to see this degenerate into trying to spec 1MB max frame size, auto-negotiation of max framesize, etc.

Both my position and that of the customers I deal with is that higher speeds, 10gbps included, is more efficient with the larger packets.

no argument from me on this one, although the actual efficiency gained comparing 1.5 K frames to 9 K frames is only ~ 2%, but when you're on the cutting/bleeding edge of technology, every bit counts.



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.

My question is not wether to support jumbos, because we all already do ... my question is should we finally spec it out?  I think we should, at a minimum, provision for it.



Michael J. Bennett
Sr. Network Engineer
LBLnet Services Group
Lawrence Berkeley Laboratory
Tel. 510.486.7913