dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
594
share rss forum feed


rchandra
Stargate Universe fan
Premium
join:2000-11-09
14225-2105

decades of traffic management experience shows

The phone companies, or at least divisions of them, have been doing traffic management for a long, long time. To me, it's not particularly surprising T with their U-Verse and VZ with their FiOS do well in this sort of metric. Even when I was a VZ DSL subscriber in south Buffalo, at least when I tried to measure it, it seemed to me that the speeds were really stable (no noticeable dips). I know that's much less than scientific, with only a few samples. But I think knowing nothing else, this is through decades of traffic engineering experience.
--
English is a difficult enough language to interpret correctly when its rules are followed, let alone when a writer chooses not to follow those rules.

Jeopardy! replies and randomcaps REALLY suck!

Crookshanks

join:2008-02-04
Binghamton, NY
The Verizon DSL forum here is littered with people who experience peak hour slowdowns. I was one of them until I raised a stink in the direct forum and had my line re-provisioned. On one hand I can't complain real loudly because they DID fix it; one wonders how many people who aren't knowledgeable enough to complain are paying for service they aren't receiving though?

Incidentally, Verizon stopped offering anything higher than 3.0mbit/s out of my CO. It seems I ordered my 10mbit/s service right in the nick of time because it's no longer offered. The tech who came out to hook up my POTS service a few weeks ago confirmed that they did this because of bandwidth limitations.


jseymour

join:2009-12-11
Waterford, MI
reply to rchandra
said by rchandra:

The phone companies, or at least divisions of them, have been doing traffic management for a long, long time. To me, it's not particularly surprising T with their U-Verse ... [does well]...

"AT&T," U-Verse comes in 8th, behind six cable companies.

Jim


rchandra
Stargate Universe fan
Premium
join:2000-11-09
14225-2105
I guess I'm interpreting "does well" a little differently. In terms of overall throughput, yes, they're just above Clearwire, and below (on most days) under everyone else. But four of those lines are in the same general neighborhood (roughly 2200-2400) with similar variability (all curves rise and fall approximately equal amounts relative to a baseline). Two of those six have higher variability in the data, most prominently Cablevision, along with Cable One. Again, I'm talking about the variability of the connection (which I think would cause the most problems in viewing a stream), not the throughput absolute value. After all, the Y axis is "kbps," not some quality number. I would think individual subscriber preferences (what the access speed is for the service they choose) could account for at least some of the relative speed differences...although U-Verse is supposed to be zippy like FiOS. Hmmm....
--
English is a difficult enough language to interpret correctly when its rules are followed, let alone when a writer chooses not to follow those rules.

Jeopardy! replies and randomcaps REALLY suck!


Smith6612
Premium,MVM
join:2008-02-01
North Tonawanda, NY
kudos:24
Reviews:
·Verizon Online DSL
·Frontier Communi..

1 recommendation

reply to rchandra
Well, that may be the issue, but the real problem is that the DSL companies are all stuck with people seeing congestion at night in some areas, speed limitations due to distance or equipment (Remotes!), and in the case of a bunch of Verizon lines, speed issues due to equipment setup/configuration. Now look at FiOS where none of that really exists. Don't think the Telcos are traffic shaping if FiOS is really kicking the cable companies off to the side like that...

rradina

join:2000-08-08
Chesterfield, MO
reply to rchandra
I'm not sure what you are trying to say. Are you making a statement about variance or throughput or both?

If I look at the chart, some cable providers, who by implication have not been doing traffic management as long as Ma Bell, do have a higher variance than ATT U-Verse and VZ FIOS. However, the variance seem the same for Charter, Comcast and ATT/VZ.

Charter has 5.2 million customers (cannot find Internet breakout), VZ FIOS has ~4million Internet customers and Comcast has 16.7m Internet customers. If anything I would have to surmise that Comcast is tops in traffic management since they server 4 times more customers and yet their variance isn't any greater than VZ.

In terms of FIOS, their higher throughput is likely due to their minimum speed being similar to cable's mid to upper tiers.


rchandra
Stargate Universe fan
Premium
join:2000-11-09
14225-2105
variance between providers...the telcos' curves seem overall flatter (less variance from day to day) than the others. And the other statement I was trying to make could probably be better made with an example: if the throughput on any given day (X axis) goes up for Charter, it goes up approximately (but not absolutely) as much for AT&T (and Comcast, and Verizon, and...)

But that's a great point about subscriber base. I hadn't thought about that.
I stand corrected.
--
English is a difficult enough language to interpret correctly when its rules are followed, let alone when a writer chooses not to follow those rules.

Jeopardy! replies and randomcaps REALLY suck!