|
Hugh,
Re: cut through market penetration: Bring this
back up in 2-3 years and we will compare notes.
Re: preemption backward compatibility: Of course 802.3
would create mechanisms to ensure that a switch implementing pre-emption would
plug and play with one that didn't. You simply make the default mode equal to
the existing mode. I know that you know this. 802.3 did for Link Ag. It did
for OAM. It did for.... Come on Hugh, this isn't that
hard.
jonathan
Jonathan,
In line:
Jonathan Thatcher
wrote:
Hugh,
Well, I
certainly can't get on board with the idea of 40 or 100 Gb/s being cheap or
simple. At least not in the next couple of years.
I never
thought of myself as small-frame-phobic. I always thought of myself as a
lover of improved cost-performance.
You are
correct that geometry matters if you want low latency.
Regarding
your comment of this being a niche of a niche, to some of us,
being a part of a 1 B$ a year and rapidly growing niche within a 50B $ a
year niche is worthy of consideration. It doesn't especially bother me
that this might be embarrassingly small and not worthy of consideration
for the largest vendors. You (and others)
would be wise not to make assumptions about what large vendors consider worthy
of attention (or not). The reason why I classify it as a niche of a niche is
that I expect most of this large and interesting market will be satisfied by
products based on standard technologies (adapted from LAN and WAN
applications). I also expect that that there will be a significant niche
demanding higher performance (in the range discussed below) that will require
more exotic architecture. To satisfy this niche, end-station vendors will need
hardware acceleration; switch vendors may use cut-through and, as a result,
hardware will be significantly more expensive. Then there is a
niche-of-a-niche that will need faster layer 2 operation than Ethernet can
provide. I expect that such a market could use existing supercomputer-defined
interfaces or may be small enough to tolerate custom or proprietary solutions.
I do not see that the niche-of-a-niche warrants the making of a new standard
for the whole of Ethernet.
If there is any demand for preemption , then
I would expect that cut-through switching would have a significant segment of
the current market as it is a tried-and-true technology that is fully
compliant with current standards. What is the current penetration of
cut-through switches in new switch sales?
Of course
a switch implementing pre-emption would interoperate with a switch that
didn't. Really Hugh, that kind of FUD is beneath
you. Of course it won't! You would have to
define some mechanism for backward compatibility that involves discovery and
negotiation before pre-emption is used. If, for any reason, a switch were to
use preemption on an interface connected to a switch which doesn't understand
preemption then the receiving switch would see a jumbled frames. At best this
would lead to packet loss, at worst it would cause a very high false packet
acceptance rate. I would expect that such a mix of PCS capabilities introduced
into the market would generate a far worse number of user issues than simply
adding (or changing) a protocol frame.
Hugh.
jonathan
Jonathan,
I don't know
why you're so scared of smaller frames - anyone who wants smaller latency
should prefer smaller frames. If you reduce the MTU to 500 bytes (not to
48) the equation swings in favor of existing standards:
6 x 500 x
8/10k + 0.5 = 2.9 vs 0.8 - still a 3x improvement using preemption, but
getting closer. Bear in mind that this is an extreme worst case
comparison. Averages will be almost identical because the preempting
packet can arrive at any time during the preempted frame; the preempted
frame might not be of maximum size; the link may be idle when the
preempting packet arrives; plus of course the packet in progress may be a
high priority packet also.
Of course if we start adding in more
delays the difference gets yet smaller (both delays increase similarly).
e.g.
Your example allows only 15m per link, you will start to run
into geometrical problems if you want to aggregate very large numbers of
nodes with only 15m per link. If there are fewer nodes then you need to
re-architect you interconnect matrix because 6 hops should be able to
accommodate many thousands of end stations.
Your example must be
assuming very aggressive cut-through switch architecture (cut through has
lost popularity in recent years, shame). If you want to conform to the
requirements of bridging then you should wait for both the source &
dest MAC address to be received before you transmit (unless you are a
repeater!). Since you are advocating preemption, I would also assume that
you must wait for the COS/TOS tag. That extra 10 bytes will be difficult
to avoid. Of course, if you decide that the error propagation of
cut-through makes the technique unfavorable then you have a full 64 bytes
of latency to wait for the CRC of the incoming frame.
Regarding
jumbo frames and complexity of end station devices, I would expect that
any device capable of filling a 10Gbps pipe will require some
hardware acceleration. For hardware implementations there really is no
significant difference between encapsulating 1500 byte frames vs 500 byte
(or even smaller) frames. Hardware which performs this high speed
operation has the advantage that it is seamlessly compatible with any
other equipment that might be connected to it. On the other hand, if a
switch started using a preemption mechanism when connected to any existing
hardware then it could be anybody's guess what would result. My assertion
is that a small reduction in MTU for the local network will yield results
which are close enough to your extreme examples to make the applicable
space where a new standard is demanded very small indeed. As I said, it's
a niche of a niche.
Better to spend our effort on cheap and simple
40Gig (or even 100Gig) and make this whole argument moot (yes, at 100G the
max length frame can be stored in 25m of
wire).
Hugh.
|