[Int-area] Re: IP Parcels and Advanced Jumbos (AJs)

There are places where we see a possible fit for some of this that pertain to special-use Internetworks that may or may not connect to what we think of as the global public Internet. Space-based networking, civil aviation, tactical networks, data centers, others but as you say with the Internet to follow. We recently gave talks on this subject at the IEEE SMC-IT STINT workshop and Netdev 0x18 conferences that drew good discussion. So, we do see some passion for the subject.

From: Robinson, Herbie <[email protected]>
Sent: Tuesday, September 24, 2024 4:09 PM
To: Paul Vixie <[email protected]>; Templin (US), Fred L <[email protected]>; Hesham ElBakoury <[email protected]>
Cc: Internet Area <[email protected]>; IPv6 List <[email protected]>
Subject: RE: [Int-area] Re: IP Parcels and Advanced Jumbos (AJs)

I think what we need to see is an overall view of how this is going to expand MTU across the Internet or at least at first in intranets (with the Internet to follow), not specs for little pieces of it.  That and some indication that there is buy-in for this among hardware vendors.  This has been a glaring problem for decades, but no hardware vendor seems to want to go first.


From: Paul Vixie <[email protected]<mailto:[email protected]>>
Sent: Tuesday, September 24, 2024 6:53 PM
To: Templin (US), Fred L <[email protected]<mailto:[email protected]>>; Hesham ElBakoury <[email protected]<mailto:[email protected]>>
Cc: Paul Vixie <[email protected]<mailto:[email protected]>>; Internet Area <[email protected]<mailto:[email protected]>>; IPv6 List <[email protected]<mailto:[email protected]>>
Subject: [Int-area] Re: IP Parcels and Advanced Jumbos (AJs)

You don't often get email from [email protected]<mailto:[email protected]>. Learn why this is important<https://aka.ms/LearnAboutSenderIdentification>
SMART Modular Security Checkpoint: External email. Please make sure you trust this source before clicking links or opening attachments.
Perhaps a plenary slot in Dublin where we can make the case?

p vixie

On Sep 24, 2024 17:00, Hesham ElBakoury <[email protected]<mailto:[email protected]>> wrote:
Hi Fred,
My impression is that this topic has not received strong support in IETF. If this is true, then how you plan to change IETF position?

Thanks
Hesham

On Tue, Sep 24, 2024, 1:44 PM Templin (US), Fred L <[email protected]<mailto:[email protected]>> wrote:
Thanks Paul, as you know well the Internet has come a long way since the days of FDDI but has
not done very well at accommodating packet size diversity. We can and should do better, IMO.

Fred

From: Paul Vixie <[email protected]<mailto:[email protected]>>
Sent: Tuesday, September 24, 2024 12:59 PM
To: Templin (US), Fred L <[email protected]<mailto:[email protected]>>; Internet Area <[email protected]<mailto:[email protected]>>; IPv6 List <[email protected]<mailto:[email protected]>>
Subject: Re: [Int-area] IP Parcels and Advanced Jumbos (AJs)

Something like this is long needed and will become badly needed. Every 10X of speed increase since 10mbit/sec has gone straight to PPS, whereas the speed increase from 3mbit/sec to 10mbit/sec was shared between PPS and MTU.

If every 10X has been shared between PPS and MTU, say sqrt(10) for each, our MTU would be well over 64K by now and our PPS wouldn't require dedicated NPU hardware to source, sink, and ferry those packets at link saturation levels.

Every attempt at PMTUD so far has failed but that's not an excuse to stop trying.

Thanks for driving this Fred.

p vixie

On Sep 24, 2024 14:39, "Templin (US), Fred L" <[email protected]<mailto:[email protected]>> wrote:
It has been a while since I have posted about this, and there are some updates to highlight.
See below for the IPv6 and IPv4 versions of “IP Parcels and Advanced Jumbos (AJs)”:

https://datatracker.ietf.org/doc/draft-templin-6man-parcels2/<https://url.us.m.mimecastprotect.com/s/nvtfCpYLPYtA5Nq6IPf7CGmc38/>
https://datatracker.ietf.org/doc/draft-templin-intarea-parcels2/<https://url.us.m.mimecastprotect.com/s/22qtCqxL9xSX9PnwHXh8CEu4yp/>

The documents acknowledge that parcels are analogous to Generic Segment/Receive Offload
(GSO/GRO) but taken to the ultimate aspiration of encapsulating multi-segment buffers in
{TCP/UDP}/IP headers for transmission over parcel-capable network paths. They further give
a name to the multi-segment buffers used by GSO/GRO, suggesting that they be called
“parcel buffers” or simply “parcels”.

AJs are simply single-segment parcels that can range in size from very small to very large.
They differ from ordinary jumbograms in several important ways, most notably in terms
of integrity verification and error correction. They also suggest a new link service model
that defers integrity checks to the end systems where bad data can be discarded while
good data can be accepted as an end-to-end function, reducing retransmissions.

Together, these documents cover all possible packet sizes and configurations that may
be necessary both in the near term and for the foreseeable future for Internetworking
performance maximization . Comments on the list(s) are welcome.

Fred Templin
_______________________________________________
Int-area mailing list -- [email protected]<mailto:[email protected]>
To unsubscribe send an email to [email protected]<mailto:[email protected]>

_______________________________________________
Int-area mailing list -- [email protected]
To unsubscribe send an email to [email protected]