Re: [ippm] [tsvwg] [iccrg] New Internet Draft: Congestion Signaling (CSIG)

[Jai Kumar <[e-mail]>]

Tom,

AI/ML and HPC clusters are evolving for multi-tenancy and will take some
time but there are other use cases for DC traffic engineering that are
captured in the draft. In such cases packets may undergo an unspecified
number of tunneled domains and as I mentioned IOAM doesn't work well when
it traverses the hierarchy of tunnels besides you have got a proposal only
for IPv6.

I have nothing against IOAM if there is an adoption and customer ask. There
are multitudes of IETF RFC drafts that are just drafts. All I am saying is
that I have spent time and  energy supporting IOAM in HW that no customer
is asking for it. You may not like it but that's the fact. Do some digging
around and find out the number of switches supporting it.

Best,
-Jai




On Tue, Feb 20, 2024 at 3:24 PM Tom Herbert <[e-mail]> wrote:

> On Tue, Feb 20, 2024 at 2:23 PM Jai Kumar <[e-mail]> wrote:
> >
> > Tom,
> >
> > Please see inline ...
> >
> > On Tue, Feb 20, 2024 at 1:54 PM Tom Herbert <[e-mail]> wrote:
> >>
> >> On Tue, Feb 20, 2024 at 1:02 PM Jai Kumar <[e-mail]>
> wrote:
> >> >
> >> > Thank you for a lively discussion.
> >> >
> >> > I just want to add a few more points in favor of using CSIG in L2.
> >> >
> >> > As we are trending towards high performance ethernet fabric for HPC
> and AI/ML clusters, there are innovations happening in both link layer and
> layer 3 headers. Amongst many some of the key requirements are
> >> > - minimal and constant overhead (in CSIG it is a fixed 16bit TPID,
> where as for IOAM IPv4 packets if I recall it is a GRE header encapsulation
> and for v6 it is Hop by Hop option)
> >>
> >> Please see the solution I proposed (again not IOAM). This could be
> >> done effectively as a fixed size header consisting of 14 byte Ethernet
> >> header, 20 byte IPv6 header, and 8 byte HBH Options containing a CSIG
> >> option. As I mentioned, hardware can identify these headers from
> >> fields as fixed offsets, and otherwise assume that all the fields that
> >> need to be processed are at fixed offsets in the packet.
> >>
> > [Jai] Proposal is good only for IPv6. What about IPv4 and/or evolving
> compressed IP headers?
> >
> >> > - independence from layer 3 transport. So that it can work in legacy
> IP networks or in optimized networks where there MAY be some form of
> compressed headers present.
> >>
> >> By legacy, I assume you mean IPv4? There are some approaches that
> >> could be applied (draft-herbert-ipv4-eh or maybe a lightweight
> >> encapsulation protocol).
> >
> > [Jai] We have a proposal. Why not look into what we have proposed? It
> works for any kind of Layer 3; IPv4, IPv6 and/or new compressed IP headers
> and even for MPLS. I do not see any good reason to invent yet another
> lightweight encap when we have CSIG as an L2 header. Also note that today's
> DC traffic is almost 80% IPv4. Having just a solution for IPv6 itself is a
> non-starter.
> >>
> >>
> >> Also, while independence for L3 transport might be a goal, as I
> >> already pointed out there are at most only two L3 protocols we need to
> >> worry about: IPv6 and IPv4. If the information is L2 then we have to
> >> map that into the different L2 technologies-- so IMO, avoiding
> >> dependence on L2 would be a better goal to constrain complexity of the
> >> solution. This is also true for the upper layer >=L4. Right now the
> >> draft is dependent on L4 or higher to reflect the signal. If we put
> >> this information in L3, then we again only have to worry how to do
> >> this for the two IP protocols.
> >>
> >> > - low latency hw implementation. AI/ML clusters are latency sensitive
> for inference and demands <.20us switch latency.
> >>
> >> Please look at the format I proposed. I don't offhand see anything
> >> that would make processing any slower than putting the information in
> >> L2. Again, everything needed could be at fixed offsets and mapping the
> >> packet to the format for processing is a matter of comparing a few
> >> fields in the parsing buffer (and most of those fields are probably
> >> already checked for normal forwarding anyway).
> >>
> > [Jai] We can have this discussion offline but putting CSIG information
> in L3 header is very complex for HW updates and also when a CSIG packet
> enters a new tunneled domain or exits a tunnel domain, Switches do not have
> the ability to infinitely parse into packet and typically operate at a cell
> boundary of 196 Bytes to 256 Bytes. You either have to bubble up the CSIG
> header from L3 to the outer tunnel header for encap or copy it down to the
> payload header when doing tunnel decap. This is a very expensive operation
> to do in HW and MAY not be possible of the CSIG tag gets pushed beyond
> single cell boundary.
>
> But earlier you said that "hough we talk about vxlan tunnels in AI/ML
> or HPC clusters but that is still a couple of years away"? Do you see
> any issues in hardware for forwarding the packet format I proposed
> (just plain switching).
>
> >
> >> > - decoupling of congestion signaling from transport. One such example
> is the Ultra Ethernet Transport proposed in UEC. Reflection for source
> based congestion control is in purview of transport as and when needed.
> >>
> >> But reflecting the information doesn't decouple congestion signaling
> >> from transport. This is the reason why the reflected signal should be
> >> in L3.
> >>
> > [Jai] It does, There MAY not be a need of reflection and intersection
> with transport if the congestion control algorithm is receiver based.
> >
> >> > - fixed offset and constant header variance for v4 or v6 frames for
> both low latency and future proofing for any innovation happening in these
> layers. Note that these clusters are not L2 bridged domains and are mostly
> L3 routed domains.
> >>
> >> This is where I don't understand how an L2 solution can work
> >> end-to-end. If clusters are L3 routed domains then how does
> >> information in an L2 header get propagated through an L3 switch?
> >>
> > [Jai] Now you are trying to understand the concept. I am glad. CSIG tag
> is an opaque tag that is not used in forwarding but is transparently
> forwarded. Think of it as a vlan tag swap with itself (if there are no
> updates to CSIG signals) or swap to a new vlan tag (if there is an updated
> signal). Bottom line is that this tag is not used for forwarding.
> > Just to clarify the concept of opaque tags are widely used in DC
> services and most of the switch vendors support it either as 4 Byte or 8
> Byte opaque tag.
>
> Is there a standard on this?
>
> >
> >> > Though we talk about vxlan tunnels in AI/ML or HPC clusters but that
> is still a couple of years away. So the argument of variable layer 2
> headers is a weak argument.
> >>
> >> But in that argument you're assuming the use of the protocol is just
> >> in high performance AI/ML or HPC clusters. We get much better utility
> >> out of developing protocols with a variety of use cases. For instance,
> >> I think CSIG will be very high value in 6G networks which will have
> >> different characteristics and probably heterogeneous L2 in the path.
> >
> > [Jai] We need to start from somewhere.
>
> You seem to be assuming that no one has ever conceived of the idea of
> signaling congestion with more bits. That's incorrect. There have been
> a number of proposals in IETF for network to host signaling even
> beyond IOAM.
>
> > And with the explosion in AI/ML use cases, this is the burning topic
> that we are trying to solve.
>
> To be more specific, this is the burning topic that _you_ are trying
> to solve; that's not the same thing as saying this is the burning
> topic that all of IETF is trying to solve. Clearly, this solution is
> targeted towards high end data centers, and doesn't  address all the
> interesting use cases of AI/ML (especially if it's bound to L2).
>
> > Why solve something which no one cares about or there is no market or
> money for it (like IOAM, sorry couldn't stop taking a jab).
>
> I have no idea why you'd think addressing congestion in the network is
> something no one cares about. As for IOAM, while you may not like it,
> I will point out that it is published RFC with several Standards track
> documents, whereas CSIG is currently just an I-D and Experimental
> status. If you want to get any traction with CSIG in IETF, I think
> you'll want a much stronger rationalization why existing IETF
> protocols cannot be adapted for the purposes of CSIG.
>
> Tom
>
> >
> >  Best,
> > -Jai
> >>
> >>
> >> >
> >> > I think the draft makes it clear that applicability of the proposal
> is DC focussed.
> >> >
> >>
> >> Sure, a lot of great protocols start with a narrow use case :-)
> >>
> >> > Tom,
> >> > Also, there is a discussion and presentation done before talking
> about the HW complexity and merits/demerits of encapsulating a v4 packet in
> IOAM GRE header and more. It should be present in archives.
> >>
> >> Still haven't found it. Maybe I'm looking in the wrong archives?
> >>
> >> Tom
> >>
> >> >
> >> > Best,
> >> > -Jai
> >> >
> >> >
> >> >
> >> >
> >> > On Tue, Feb 20, 2024 at 12:35 PM Sebastian Moeller <[e-mail]>
> wrote:
> >> >>
> >> >> Hi Tom,
> >> >>
> >> >>
> >> >> > On 20. Feb 2024, at 21:26, Tom Herbert <[e-mail]>
> wrote:
> >> >> >
> >> >> > On Tue, Feb 20, 2024 at 12:09 PM Christian Huitema <
> [e-mail]> wrote:
> >> >> >>
> >> >> >>
> >> >> >>
> >> >> >> On 2/20/2024 9:55 AM, Sebastian Moeller wrote:
> >> >> >>>> That's more of a statement of security and not feasibility.
> There's simply no security in the Internet, so we cannot trust or validate
> that anonymous intermediate nodes are going to write correct information.
> Any plain text in a packet on the Internet is subject to inspection and
> modification if the data isn't authenticated, and in the worst case this
> could be a DoS vector by writing bad information.
> >> >> >>> [SM3] Indeed, but e.g. for TCP you would need to know a lot
> about the most recent packet to be able to play games, no? So either you
> are on path and already can drop/duplicate packets at will or you are off
> path but still need a recent enough veridical packet to be able cause
> mischief, no? (I might be insufficiently creative in attack vectors)
> >> >> >>
> >> >> >> I am analysis congestion control information using the framework
> of
> >> >> >> "honest signals". In human communication, "honest signals" are
> those
> >> >> >> that cannot be easily faked by the communicator. For example,
> smiling is
> >> >> >> not really a honest signal, because it is easy to fake; blushing,
> on the
> >> >> >> other hand, is hard to fake.
> >> >> >>
> >> >> >> When it come to Internet wide congestion control, we have pretty
> much
> >> >> >> the same issue. Networks may want to fool the application for a
> variety
> >> >> >> of reasons, and may start faking congestion signals. Some of these
> >> >> >> signals are hard to fake. End to end data rate for example:
> slowing a
> >> >> >> specific stream of packets is hard to fake; measuring the end to
> end
> >> >> >> data rate is a pretty good indication of the state of the
> network. End
> >> >> >> to end RTT is also a rather honest signal: yes, routers could put
> some
> >> >> >> specific packets in a slow queue, but that requires resource.
> >> >> >>
> >> >> >> Packet losses almost belong in that category. They are not hard
> to fake,
> >> >> >> routers could play favorites and selectively drop packets with a
> certain
> >> >> >> profile. But dropping too many packets affects the "quality
> rating" of a
> >> >> >> provider, so there is some pressure to not fake it. That pressure
> is
> >> >> >> probably one of the reasons behind bufferbloat. The main problem
> with
> >> >> >> packet loss as a signal is that losses may have other causes than
> >> >> >> congestion.
> >> >> >>
> >> >> >> ECN is not really a honest signal. Setting a bit in a packet
> header does
> >> >> >> not require a lot of efforts, so routers could do that to play
> >> >> >> favorites. In fact, past bugs in some networks caused almost all
> packets
> >> >> >> to be marked as CE. Using ECN is very nice when you can trust it,
> but
> >> >> >> end nodes should probably do that cautiously, detecting for
> example a
> >> >> >> sudden raise in ECN marks rather than reacting to an average
> value.
> >> >> >>
> >> >> >> ECN is just one bit. There is always a temptation to do a better
> ECN
> >> >> >> with many more bits. For example, CE directs a sender to slow
> down. It
> >> >> >> would be nice to have a corresponding "All clear" signal telling
> the
> >> >> >> senders that they can speed up. L4S attempts to do that by
> modulating
> >> >> >> the CE bit, so that a low frequency kinda indicates "all clear",
> while a
> >> >> >> high frequency says "slow down", and give some indication of how
> much.
> >> >> >> Suddenly, one bit becomes several bits, just spread over many
> packets.
> >> >> >>
> >> >> >> The idea of adding more bits in packet headers is not exactly new
> -- see
> >> >> >> for example TCP QUIC Start by Sally Floyd et al., RFC 4782,
> January
> >> >> >> 2007. The problem is that the more bits you add, the more you
> exacerbate
> >> >> >> issues of trust, and also risks of bugs. "Many more bits" may
> work in a
> >> >> >> controlled environment, but I really do not see that working on
> the
> >> >> >> whole Internet.
> >> >> >
> >> >> > Hi Christian,
> >> >> >
> >> >> > Do you know what the state of ECN deployment over the Internet is?
> >> >>
> >> >> [SM5] Nobody knows exactly, but quite a lot of Linux servers use the
> LInux defaults and will use ECN if the client negotiates it. In my qdisc
> statistics I routinely see not only drops, but also CE marks logged (from
> my AQM).... so believe it or not, ECN over the internet mostly works...
> >> >>
> >> >>
> >> >> > It seems to me that if someone is sending to an arbitrary host
> over the
> >> >> > Internet they're already pretty much accepting "best effort"
> service:
> >> >>
> >> >> [SM5] Not sure about the US, but that is all ISPs offer to end users
> over here... but ECN works well even over a best effort internet access
> link in my personal experience.
> >> >>
> >> >> > long latencies and with potentially high variance, such that
> getting
> >> >> > fined grained congestion information from intermediate routers,
> even
> >> >> > if it's honest, probably doesn't add much to the information that
> we
> >> >> > can derive from packet loss or measuring RTT with no additional
> >> >> > mechanisms or implementation.
> >> >>
> >> >> [SM5] How can you come to that conclusion without ever trying?
> >> >>
> >> >>
> >> >> > The situation is very different in a limited domain which could
> >> >> > include large service provider networks.
> >> >>
> >> >> [SM5] Indeed, papers discussion 'better congestion signalling' often
> come out of those environments. But IMHO not because these methods only
> help in those environments, but that this is where people are willing to
> spend the money to test and implement potential solutions.
> >> >>
> >> >> Sebastian
> >> >>
> >> >> > In that case more information
> >> >> > is good, it's easier to provide security so we can trust the
> >> >> > information, and we're not restricted to just one or two bits of
> >> >> > information to carry the information in a packet. This is also
> where I
> >> >> > see host-to-network signaling being useful-- this allows
> applications
> >> >> > to request QoS for their packets
> >> >>
> >> >> >
> >> >> > Tom
> >> >> >
> >> >> >>
> >> >> >> -- Christian Huitema
> >> >>
> >> >>
> >> >
> >> > This electronic communication and the information and any files
> transmitted with it, or attached to it, are confidential and are intended
> solely for the use of the individual or entity to whom it is addressed and
> may contain information that is confidential, legally privileged, protected
> by privacy laws, or otherwise restricted from disclosure to anyone else. If
> you are not the intended recipient or the person responsible for delivering
> the e-mail to the intended recipient, you are hereby notified that any use,
> copying, distributing, dissemination, forwarding, printing, or copying of
> this e-mail is strictly prohibited. If you received this e-mail in error,
> please return the e-mail to the sender, delete it from your computer, and
> destroy any printed copy of it.
> >
> >
> > This electronic communication and the information and any files
> transmitted with it, or attached to it, are confidential and are intended
> solely for the use of the individual or entity to whom it is addressed and
> may contain information that is confidential, legally privileged, protected
> by privacy laws, or otherwise restricted from disclosure to anyone else. If
> you are not the intended recipient or the person responsible for delivering
> the e-mail to the intended recipient, you are hereby notified that any use,
> copying, distributing, dissemination, forwarding, printing, or copying of
> this e-mail is strictly prohibited. If you received this e-mail in error,
> please return the e-mail to the sender, delete it from your computer, and
> destroy any printed copy of it.
>

-- 
This electronic communication and the information and any files transmitted 
with it, or attached to it, are confidential and are intended solely for 
the use of the individual or entity to whom it is addressed and may contain 
information that is confidential, legally privileged, protected by privacy 
laws, or otherwise restricted from disclosure to anyone else. If you are 
not the intended recipient or the person responsible for delivering the 
e-mail to the intended recipient, you are hereby notified that any use, 
copying, distributing, dissemination, forwarding, printing, or copying of 
this e-mail is strictly prohibited. If you received this e-mail in error, 
please return the e-mail to the sender, delete it from your computer, and 
destroy any printed copy of it.