🚀 go-pugleaf

[POSTED TO comp.dcom.modems.cable; PLEASE REPLY THERE]

In <3A7C2957.6083F801@mynetwatchman.com>, Lawrence Baldwin
<baldwinlSPAMMENOT@mynetwatchman.com> wrote:

>> > I don't think so.  Upstream oversubscription is really an issue only on
>> > cable, since (1) DSL has dedicated upstream to the DSLAM, and (2)
>> > backhaul links are symmetric (as compared to asymmetric subscriber
>> > connections).
>
>I disagree on one finer point.
>
>This is where I think we agree:
>
>* TCP itself will limit downstream throughput when the upstream channel
>becomes saturated

Agreed.  However, this is only a concern on *asymmetric* links where the
downstream capacity is substantially greater than the upstream capacity.

>* Cable upstream saturation (between customer and CMTS) is highly likely
>because of the shared nature of the link

I don't think it's necessarily any more likely -- that would depend on
many complex factors (activity patterns, channel loading, etc.).

>User A can be degraded because B, C, D, and E are uploading at max rate.
>-or-
>User A can be degradged because just the cumulative affect of all users
>ACK traffic causes saturation.  (this would seem to occur only where
>*many* users share the same local segment).

On cable, A-Z can all be greatly degraded because of heavy (rather than
typical) upstream use by Y and Z, excessive demand for available
upstream by A-X, or some combination of both.

>* ADSL upstream saturation (between the customer and the DSLAM) is
>highly
>unlikely because it is a private link. (Though the customer could
>saturate it himself with some kind of outbound datastream...e.g. UDP
>broadcast video).

Again, not necessarily.  What's different is that on ADSL, A-X are
unaffected by any amount of upstream use by Y and Z -- only Y and Z are
affected, each only by its own activity.

>It's NOT going to saturate just due to his outbound ACK traffic.
>Assuming download rate of 150,000Bytes/sec, with 1500 bytes
>packets...that's about 100packets/sec.  That should result in upstream
>50packets/sec of 64byte ACKs...that's 25Kbps, far less than the 256Kbps
>typical ADSL upstream bandwidth.

FTP downloads certainly won't saturate the uplink.  However, HTTP is
more problematic -- I've measured 1:7 uplink to downlink ratios (bytes,
not packets) on popular websites.

>Here is where I disagree:
>
>Downstream degradation occurs WHENEVER the uplink is saturated.  Granted
>the more limited (and shared) upstream bandwidth of cable makes this an
>easy choke point.  However, ANY uplink that saturates can have the same
>affect.

Again, this is only a concern on *asymmetric* links.  Saturation of the
uplink on a *symmetric* path will not have an adverse effect on the
downlink.

>Even though bandwidth between the DSLAM and aggregation router is
>assymetric (e.g. 45Mbps full-duplex),

Typically *symmetric*.

>you could still saturate upstream
>before downstream, depending on what users are doing.

Possible, but unlikely given typical usage patterns of far more download
than upload.  But so what?  Saturating the upstream will not adversely
affect the downstream on a *symmetric* link.

>For example, if 5% (175) of the ~4000 ADSL users aggregated into a DS3
>ran active web servers (uploading at constant 256Kbps), you'd saturate
>the upstream.

I'd say it's typically more like 2000-3000 subscribers per DS3, with
uplink capped at 128 Kbps, so that it would take more like about 15%.
But again, so what?

>I'm not saying that this is extremely likely, but I
>believe the best way to quantify the probablility is to quantify:
>upstream oversubscription rate

With all due respect, I don't think it's that simple, since there are so
many complex factors at work.  Regardless, the backhaul issue is
essentially the same for both ADSL and cable (beyond the DSLAM and the
CMTS respectively), and since the backhaul is typically *symmetric*,
it's not subject to the TCP/IP asymmetry problem.

>To do an apples to apples comparison, you need to compare both
>technologies at the same layers: Customer to Local Aggregation (I think
>we agree cable and ADSL infrastructure upstream from the aggregation
>layer use similar architectures).
>
>ADSL: Subscriber to Aggregation router
>Cable: Subscriber to CMTS router

ADSL:  Subscriber to DSLAM (not beyond)
Cable: Subscriber to CMTS

The backhaul beyond the DSLAM or CMTS is essentially the same.

>I suspect the upstream oversubscription rate (at the local aggregation
>layer) of cable is 10x-20x more than ASDL.  I think this is a more
>generic way to quantify the probability of upstream saturation and
>provide metric to compare different technologies on equal ground.

I'm sorry, but I don't see the logic -- I think that's like comparing
apples to oranges.  One (cable) is shared; the other (ADSL) is
dedicated.

>In fact we already know that the way to compare different ISPs to each
>other is to understand their downstream oversubscription policies.  What
>I find intriguing, is that you need to look at upstream as well...this
>is a real eye-opener for me.

You do on cable, but not for ADSL.  How can it be possible to
"oversubscribe" a dedicated link?  By definition the subscription rate
on the ADSL link (not the backhaul beyond the DSLAM) is 1:1.

>Taking this concept a bit further, my real suspicion is the
>oversubscription at the *backbone* layer is what is causing the most
>differences in experienced throughput variances between different
>providers (whether ADSL or cable, or whatever).  Wish I had the data to
>back that up though.  

With all due respect, I think you're reaching invalid conclusions by
extending an invalid comparison.  I think the big issue is the serious
adverse impact of saturation of the shared uplink on cable, a problem
that doesn't affect ADSL, even though both are asymmetric.

>Upstream saturation might actually be an even bigger issue at backbone
>interconnect layer.  Think about it, how many ISPs merge ADSL traffic
>(downstream traffic) with commercial T1 traffic (primarily upstream) at
>the backbone layer.

That could be an issue, but again, since the pipe is typically
symmetric, it's not subject to the TCP/IP asymmetry problem.  Simple
pipe congestion is completely different and well understood -- let's not
confuse the two different behaviors.

>I suspect that few considered the impacts of upstream saturation on ADSL
>throughput...or ADSL downstream traffic saturating the reverse channel
>of their T1 customers...hmmmmm.

I've not seen any real evidence of a significant problem in this area --
have you?  Regardless, the issue would be the same for cable as for
ADSL.

-- 
Best regards,
John Navas     <http://navasgrp.home.att.net/>
     CABLE MODEM/DSL GUIDE:  <http://Cable-DSL.home.att.net/>