Comcast Cheating On Bandwidth Testing?

dynamo52 writes "I'm a freelance network admin serving mainly small business clients. Over the last few months, I have noticed that any time I run any type of bandwidth testing for clients with Comcast accounts, the results have been amazingly fast — with some connections, Speakeasy will report up to 15 Mbps down and 4 Mbps up. Of course, clients get nowhere near this performance in everyday usage. (This can be quite annoying when trying to determine whether a client needs to switch over to a T1 or if their current ISP will suffice.) Upon further investigation, it appears that Comcast is delivering this bandwidth only for a few seconds after any new request and it is immediately throttled down. Doing a download and upload test using a significantly large file (100+ MB) yields results more in line with everyday usage experience, usually about 1.2 Mbps down and about 250 Kbps up (but it varies). Is there any valid reason why Comcast would front-load transfers in this way, or is it merely an effort to prevent end-users from being able to assess their bandwidth accurately? Does anybody know of other ISPs using similar practices?"

Easy

[RalphSleigh]

Speed up web browsing for their customers while keeping those dirty bittorrent pirates at bay?

Re:Powerboost

[kvezach]

Sounds like that could be tricked... instead of downloading a single ISO file, download 35 20MB files at boosted speed. Write a script to automate it, even. Or am I wrong here? If they disregard connections and turn off the boost after 20MB from when you first connected, then just downloading 21MB and disregarding the results for the first 20 should return the correct results for bandwidth tests.

It gives priority to interactive tasks

[Anonymous Coward]

The total bandwidth for all ISP's is limited and people moving large files can dramatically affect the experience of interactive users. Give priority to the user who is clicking on a lot of small web pages and they will get better response. Non-interactive tasks like downloading don't need that kind of response.

Re:This is an advertised feature I believe

[Ed Avis]

So... how can you tweak your Bittorrent client to fool Comcast into thinking it is making lots of small downloads?

may be there is no malice here?

[Anonymous Coward]

If what you say is true, there is a possibility comcast is doing it with good intentions. In mainframes, we have a concept of "period performance objective", the main idea being transactions which complete in short intervals get priority over transactions that take longer. Every transaction (in a group) will start with same priority, if it doesnt complete by a predefined period, its priority gets lowered a notch. if the transactions didnt complete in the next period too, the priority gets lowered even further. the period length, number of periods, and the quantum of reductions are all customizable. This is usually defined for online users. so a user executing a transaction that takes 0.1sec will keep getting faster response, whereas a user executing a 10sec transactions will get sluggish response. so, comcast may be doing the same. it is protecting ordinary browse or chat or email from heavy downloaders.

Re:This is an advertised feature I believe

[xSauronx]

This is it, and it should be amazingly obvious to any "sysadmin" who should know something about general browsing habits. I worked for a wISP for a year and this was a standard feature offered by the company for no extra charge. Max subscriber speeds were 1.5 Mbps, but for about 20 seconds ALL traffic was burst to 2 Mbps, regardless of the subscriber paying for the 500 k/s speed or something higher. For general browsing and light email, it made all the customers quite happy to have things terrifically speedy.

Re:This is an advertised feature I believe

[NickCatal]

There is a 'cooldown' timer. You need to be at very low bandwidth levels for a specific amount of time (or some other measure) before you can 'burst' again.

In chicago it is 12mbps then down to 6 or 8 depending on your plan. To do a proper speedtest on comcast you need to download a 100-200MB file. Although if you are getting 12mbps easily odds are you are getting your rated line speed.

Re:Gasp!

[dal20402]

The usual Slashdot "assume dishonesty before checking out the facts" attitude...

Except that they only advertise 8Mbps sustained speed, which is what you get. They also advertise PowerBoost, which gets you ~25Mbps for a few seconds.

Comcast needs to be drawn and quartered over their forged packets, but they haven't done anything dishonest in advertising their speeds, at least not where I live. I do indeed get >20MBps for a few seconds and then 8MBps until the cows come home.

Earthlink Cheats with Latency too

[goombah99]

I've noticed that my earthlink connection gets good marks for speed on tests but the speed tests take many seconds to actually start. In deed all my pages seem to load with multiple seconds of latency then all at once. (no it's not the browser, I have multiple computers that I move between home and work so I can do test across different ISPs)

I think what they are doing is giving me 1000KBs at periodic intervals or with a high latency such that my peak speed is high but my average speed is low. My latency for 600 miles connections is a good fraction of a second. Pages that have lots of element load hideously slowly compared to pages with a big download, presumably because I'm paying this huge latency penalty multiple times.

Re:Makes Web Browsing Seem Faster

[Shakrai]

Reliable streaming depends on a constistent rate.

Sure about that? I've watched how Netflix instant view uses my connection -- it pegs it for about 20-30 seconds at a time (8Mbits) and then idles for a period before downloading again. It never maintains a constant transfer, except for the first two or three minutes of watching (presumably filling up the buffer?).

In the end it averages out to around 2.2Mbps, but the actual transfers themselves usually come in bursts.

Re:This is an advertised feature I believe

[Crazy Taco]

Doesn't Comcast advertise this "SpeedBoost" as a feature - the language in their ads is something like "get massive super speed for the first 10MB of a download, then it will revert to your provisioned line speed"... So, it actually *is* a good thing rather than something to pad bandwidth tests, and it does generally help your general user, right?

Except that I never get more than my apportioned amount. In other words, my SpeedBoost never goes faster than the 6MB I actually pay for. I think that's what the person who wrote the article is saying too: "Goes at the speed they paid for, which is really fast, for a short time period and then drops to something like 1.2 MB, which is clearly slower than most comcast plans."

Re:This is an advertised feature I believe

[croddy]

I've made something of a game out of it, actually. With careful tactics, one can easily hit as much as 1.0 MiB/s upstream for short periods. I use Deluge to play. My present record is 2.4 MiB/s, on an Ubuntu 7.10 torrent for which I already had all the file data.

First, configure your torrent client to use a modest number of connections -- limit it to, say, 250 connections globally and 70% of your nominal upstream speed. Then, get on a very large, active torrent and build up a few minutes' worth of downloaded data. Once you're in the swarm, open everything wide up -- no global connection limit, no bandwidth cap, and no per-torrent upload slot limit. If your client has a bandwidth chart, watch it scroll by and enjoy the thrill as your upstream bandwidth surges to heights like you have never seen before. Of course, eventually the Power Boost will wear off and some connections will finish as their pieces are completely transferred, but it's fun while it lasts.

Re:This is an advertised feature I believe

[CharlieHedlin]

They aren't shaping based on individual downloads. They are going to shape all the traffic going to your cable modem as a single stream. There may be other outbound queues within the shaper to provide fair queuing and such, but that gets too complicated for this post.

Most shapers (including the ones in their broadband routers) allow a variety of parameters.
You can set a sustained rate, peek rate, and burst size. For example a common implementation would have the following values (I haven't worked with Cisco QOS much, so it may be implemented differently but the principals are the same):
sustained rate: 2mbps
burst rate: 10mbps
Max burst size: 10 Megabytes

The burst size counter is depleted as you download over 2mbps, and replenished when you download under 2mbps.
When you download a 100MB file you will deplete the burst size at 8mbps. You can download at 10mbps for 10.5 seconds, at which point your download will drop to 2mbps and will stay there until you slow your transfer rate. If you stop downloading completely it will take 42 seconds to refill your burst counter.

It messes up Netflix Watch Instantly

[ukemike]

This would explain why Netflix's Watch Instantly feature always stops after a few minutes and has to re-buffer with the message "your internet connection has slowed." It's really irritating. It's been doing ever since Comcast dropped my monthly rate and told me they were signing me up for "faster" service. !@#$%@#$^%

Re:It's true

[Anonymous Coward]

I've never had such poor overall bandwidth utilization until I had to get Comcast (it's the only fast game in town where I am).

The second--and I mean the absolute second--FIOS is available in my area, I'm taking a shit in a box, putting my Comcast cable modem in with it, and sending it back to Comcast: "I've put up with your guys' shit long enough--it's time you put up with some of mine".

Re:This is an advertised feature I believe

[eviljames]

I use Shaw, have the "Extreme" 10mbps package, and get a pretty consistent 10mbps. The powerboost maxes out the capacity of their cable modems (25mbps) for 5-20 seconds and drops you back to your prescribed speed. In BC you can buy Shaw Nitro which puts you at 25mbps all the time. What Comcast is doing sounds more like selling the service at 10mbps, throttling their customers to 1.5 for every day browsing, and using this same PowerBoost when they run bandwidth tests. Which, ultimately, is fraud imho.

Re:PowerBoost uses a 30 second average, not filesi

[arodland]

Yeah, I did some testing on my own a while back and my theory is that it's a token-bucket sort of thing, implemented in the modem. Whenever your aggregate bandwidth is less than X for more than a certain amount of time, it allocates a "token" and resets your cap to 2*rated. The longer your connection is non-busy, the more tokens you get, up to a certain point (when the bucket is full). Then, when you start moving some data, and you go over your rated limit (which, after all, is half of what the modem is giving you), it starts taking tokens out of the bucket. When the bucket is empty, it re-caps you at your rated speed, and no more boost until you start collecting tokens again, which means a period of inactivity.

And yes, as the other commenter pointed out, this is actually an entirely sensible way to deal with "bursty" internet use and improve user experience without actually buying any more bandwidth. It would be really sweet if Comcast didn't do other stupid shit ;)

Not hard to get fully advertised bandwidth les 01%

[IBitOBear]

For all this crying, I found it very easy to get my full advertised rate from Comcast. I use a Linux box and throttle my upstream flow so as never to lose a TCP ACK in the virtually non-existent transmit buffer in my cable modem. Never lose an ACK means never suffer from TCP throttling.

Since TCP accelerates linearly and falls back exponentially, each fallback is disastrous, and if you fall back once, you are likely to do it several times in a row from the same cause.

So find out what your upstream speed is _supposed_ to be and throttle your output to between 98 and 99 percent of that value and you will get a good 7.8mbps sustained download rates.

Of course the average consumer doesn't know how to do this, but that isn't Comcast's fault. It's just wasted bandwidth.