| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
|
(cross posted to
STDS-802-3-NGECDC@xxxxxxxxxxxxxxxxx &STDS-802-3-10SPE@xxxxxxxxxxxxxxxxx) Thanks for reading. I’m the minute taker (should that be meeting notes recorded) in this case. The specific note (for those who have not read the minutes yet) says
Reduced Minimum Frame Size David Brandt Rockwell Automation
Presenter reviewed material regarding frame content use cases.
Presenter’s conclusion, ROI of changing MAC to reduce frame size questionable, does not see compelling reason to address this.
Individuals voiced opinions regarding whether a change in frame size would be a good idea in this project, including:
o Reducing frame size could introduce compatibility and network interconnection issues.
o It’s beyond the scope of a PHY project, impacts 802.3 MAC, and also impacts 802.1.
I believe that I captured the concern correctly as expressed on the call. If we just talk about not padding up to 64 bytes, I agree with you. If we consider trying to reduce the frame size to make it really small, maybe make assumptions like we can’t have VLAN tags, or MACSec then I think that is potentially
of interest to 802.1. Does 802.1AC/Q today make any assumption about the smallest legal minimum MSDU size that can be provided across 6.6 Internal Sublayer Service (ISS)? Regards Peter _______________________________________________
Peter Jones
Chair NBASE-T Alliance Cell: +1 408 315 8024 Twitter: @NBASETAlliance _______________________________________________
From: Pat Thaler [mailto:000006d722d423ba-dmarc-request@xxxxxxxx]
Sorry I couldn't make the first call. I agree with the conclusion in the minutes on David Brandt's presentation on Reduced Minimum Frame Size but I notice an inaccuracy in one of the comments recorded in the minutes; i.e., "It’s beyond the scope of a PHY project, impacts 802.3 MAC, and also impacts 802.1. " "also impacts 802.1" isn't accurate. IEEE 802.1 standards such as IEEE Std 802.1Q have to work with a variety of IEEE 802 MACs and some, e.g. IEEE 802.11, don't have padding to achieve a minimum frame size. That was true of 802.4 and 802.5
as well so there never has been a higher layer assumption of a minimum frame size. Reducing minimum frame size shouldn't affect the assumptions of 802.1 standards. (Of course, even though it doesn't affect the standards directly, it could affect implementations
of those standards as they may have architectural assumptions based on knowing that Ethernet has a minimum frame size.) Regards, Pat On Tue, Aug 23, 2016 at 2:28 PM, George Zimmerman <george@xxxxxxxxxxxxxxxxxxxx> wrote:
|