Slashdot Mirror
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?"
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?
This is because of powerboost. As I understand it, powerboost makes the first 20MB download at a higher rate than your advertised bandwidth. Since bandwidth tests are done on such small files, you get a worthless result. The idea is that people who download lotsa of relatively small files get better performance, where as people downloading a lot of huge files like ISO images, full length movies, et cetera willg et initially good speed but after 20MB will feel like they are getting gipped.
Derek Greene
Seriously now, is _anyone_ surprised over _anything_ bad Comcast does to their customers anymore?
Most traffic is HTTP, being very small files. If it starts off very quickly, most web browsing would go extremely fast, while larger files would go at "normal speed"
Speed up web browsing for their customers while keeping those dirty bittorrent pirates at bay?
Come as you are, do what you must, be who you will.
What you're seeing is expected behavior. Comcast's SpeedBoost technology maxes out your connection for an arbitrary number of seconds at the begining of a transfer, and then reverts to a normal speed. If you watch their commercials advertising SpeedBoost closely you will see the disclaimer at the bottom of the screen.
The majority of the downloads would be for web pages which are pretty small. I would think that's the reason they do this as it would make the web browsing experience seem faster.
Comcast? Dishonest? Say it ain't so!
All kidding aside, this wouldn't surprise me too much. Comcast (and probably all other providers) are advertising this super-mega-intarweb speed as "up to x mbps." So, theoretically, as long as *one* site can provide data at that rate, their marketing garbage still stands. Even if 99.9% of the other websites top out at 4kbps, if Speakeasy's speed test says it can transfer a file at 15mpbs, technically Comcast is correct. They are giving you "up to 15mbps."
Proudly supporting the Libertarian Party.
Most internet browsing is with relatively small amounts of data, so wouldn't front-loading of this nature noticeably increase browsing performance? Since this kind of performance is noticed by the majority of users it would seem to be something that increases their perception of their connections' speed.
I'm not saying that Comcast might not be cheating on purpose for speed tests, I just think that there might be another reason behind it other than just to make their test scores artifically high.
Sounds like they have simply optimised their network to favour "bursty" usage, for example web browsing. This would seem a sensible thing for a consumer ISP to do.
Is there any valid reason why Comcast would front-load transfers in this way
Your average webpage is not 100+MB. If they give you full bandwidth for, say, 2 seconds - most reasonable webpages will download completely within that time. It's not "cheating" exactly since they don't guarantee those speeds, but "up to" those speeds. They're not the only ones who do it, either.
Still a sleezy thing to do...
=Smidge=
http://www6.comcast.net/powerboost/
All it does is give you short bursts of high bandwidth and is really more talk than usefulness.
My ISP, Cox, does this too, though once the "PowerBoost" thing is off, I steadily get the bandwidth I'm supposed to get. Dunno about Comcast.
Scorta futuere amo!
Some consumers may not notice the speed increase when downloading smaller files, such as text-based e-mails and simple Web sites with few graphics. However, customers who frequently download large files, such as software, games, music, photos, and videos will now download at speeds that are faster than ever before. For example, PowerBoost significantly reduces the time it takes to download a one hour television program. Comcast subscribers at the 6 Mbps tier would reduce their wait time in half - from 4 minutes and 29 seconds to 2 minutes and 15 seconds. And MP3 fans will be able to download music files as fast as 2.2 seconds! See more here
Full Tilt
Um, this isn't a new concept, nor is it particularly sneaky:
http://en.wikipedia.org/wiki/Token_bucket
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.
Think typical internet usage, its bursty. For more typical user case of relatively small files you get a nice boost in performance. If you download something bit then its another matter. The quick gives improves web surfer latency without taking much of the bandwith. They have limited bandwith and they probably cannot give that kind of speed up all the time.
©God
Oh come on people. Can you please just stop the witch hunt? Comcast advertises this as a feature, and to me it's actually pretty convenient. I can imagine that most downloads are probably less than 10MB in a single burst, so giving the user the first 5 seconds or so of transfer at a higher rate helps the consumer see a faster internet. Keeping sustained transfers at the advertised cap speed ensures one user can't gobble up all the bandwidth over a prolonged period of time. Everybody wins.
This is not too different than how metered T/OC-x connections are operated. These metered T/OC-x connections bill you based on your sustained bandwidth, so yeah, you can have short bursts without getting a gigantic bill, but if you are pegged at 50% all the time, your bill will be huge.
"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?"
The thing I'd like to know is how you all make it through the day seeing conspiracies and evil doers behind every bush? I'd think you'd burnout.
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.
I've come to the conclusion that those Speakeasy tests are way too optimistic on my RoadRunner connection as well. Need to find a more reliable way of testing. I wouldn't be surprised if ISPs just simply boost connections to Speakeasy as well -- would at least explain how when browsing all other sites is slow, and then doing the Speakesy test I get a high score.
"Thanks for all the money you paid to us. We've used it to buy off ISO among other things" -Microsoft
The only way to determine speeds is by torrent because any amount of latency will reduce bandwidth over TCP, so to compensate for that we need to connect to as many fast clients as possible, that are as close as possible to you. I've gotten 24150kbps down and 2050kbps up on torrent.
Oh, but we're talking about the American ISP Comcast.. that throttles everything.
If you are only getting 1.2Mb sustained on Comcast you have a problem. I can pull 6Mb steady for hours on end using Comcast. Like others have said though...Speed Boost will make tests show different numbers at times.
~Sun
You just might want to check that their connections are properly tweaked as far as RWIN, MTU, etc go. 14/5 compared to 1.2/290 is a vast, vast difference that should never happen if they are paying for a certain tier of service, even if it is advertised as "up to" that higher rate. I'd also do a smoke ping and line quality tests, etc over at BroadbandReports, because there is something definitely not right with those connections if that is the average drop in performance. There may also be mis-configured firewalls, routers, mis-provisioned lines, water leaks, etc causing such issues.
My advertised and provisioned rate via Atlantic Broadband cable is 5/512. I am actually getting closer to 6 or 7 down and 468 up at all times due to some tweaking I did. Even the AtlanticBB tech seemed a bit shocked that I was getting more than 5 down, and said it was unusual, but they wouldn't re-provision the line or anything because of it. I count myself lucky, because Verizon's service here is absolute rubbish - $25.00/month for 1.5/768 DSL that, shall I say in the politest way possible, isn't actually working for more than two weeks per month because they are too cheap to replace lines that were put up in this town sometime in the 1950's at the latest (Not to mention they never actually bother to show up for scheduled appointments to rewire buildings that were constructed pre-1900, such as mine - big old Victorian type home turned into apartments).
Powerboost does mess with speed testing, however those "tests" are very rarely accurate anyhow, as I can rate higher on a test to Seattle or Los Angeles than I do to say Pittsburgh, Toronto or NYC, which are MUCH closer to where I live (by several thousand wire miles). It's more accurate to calculate your average rates by downloading/uploading large files from/to a university/public FTP or something, at least in my experience.
@Mindless Drivel: 100% of Twitter posts ever Tweeted.
I was assuming they where in bed together, Comcast and the 2 top bandwidth speed testers that is. Though Speed boost makes just as much sense.
/. post http://ask.slashdot.org/article.pl?sid=07/07/08/2116214
Weather this is true or not, it's just a revisit to an older
Charter does this too.. It's standard practice lately for cable providers. It makes them look better and it helps out the end user by allowing them to "burst" small amounts of data (like websites). This is a function of traffic shaping, I dont see any harm in it. Just test your connection speed with real world application instead of these speedtest sites. You will get a lot better picture of how fast and "stable" your clients connection is if you download a 100mb file from an ftp site and then put it back up there. You will be able to see how much jitter they have in their connection and work out an average speed.
Slashdot taught me how to use the preview button!
Internet connections are not just for websurfing and emails anymore.
Of course it depens on the user, the average traffic from my xbox 360 alone (in gaming, demo downloads, movies etc) in one day, is more than my parents have in a month with their just light surfing and email use. And I don't use my xbox that much.
It is an issue they have to face now. Legal traffic alone these days for high tech households internet use, can pass your ISPs secret acceptable use limits.
None of the Comcast Lies and Propaganda about this so called "Powerboost" are true for me at all. In fact, last time I ran a speed test I only got 384 kbps up and something like 4-6 mbps down. Hardly an "improvement". In case you were wondering, no, I wasn't running any torrents at the time. In fact, I didn't have anything else going at the time so I have no idea why it was so slow.
that takes advantage of this by downloading in an on/off switch manner.
This is a result of "burst rate" which is an asignable property in QoS. The idea is to allow small files like web pages to load much quicker then large file transfers. Most ISP's are doing this now as a means to speed up web browsing. The best way to get an acurate speed mesurment for file transfers is to download a large file while using bandwidth monitoring software.
I know that this is slashdot but I'll try to answer some of the OP's question anyway. Of course I won't do any original research myself, but rather rely on information from the previous posters or make things up as I go.
Q1. Is there any valid reason why Comcast would front-load transfers in this way?
Yes. Most requests from browsers are for short files. By upping the speed for short requests, pages will render faster. This is a plus for the user, as he spends less time idling. Long downloads on the other time are expected to take a while to complete; the user expects to be able to walk away from the computer for a while. Thus Comcast can argue that they have greatly enhanced the experience of the web browser by stealing a few cycles from the downloader. I would welcome such a plan as long as the ISO downloading speed is reasonable.
Q2. Is it merely an effort to prevent end-users from being able to assess their bandwidth accurately?
It would have that effect on a poorly designed bandwidth test. Bandwidth testers try to make the download size long enough to counteract tcp connection costs and to average over variations in download speed. Comcast has just given them another variable to take in to account. Interestingly, there are some test suites that are designed to detect what Comcast is doing and give them extra credit for it. They bill their tests as real world throughput tests. They want to indicate what the effective bandwidth is while browsing web pages that reference many images or javascript files.
Q3. Is Comcast cheating?
If Comcast is just doing this when accessing known test sites then they are cheating. If this is their policy for all connections then the worst that can be said is that they are optimizing their service to a particular class of users (surfers as opposed to downloaders). If you are in this category, then you should be happy.
My made-up statistics show that the vast majority of users are downloading relatively small files. When you hit a web page, the browser downloads many small files. When you download anything via a P2P protocol, you're once again downloading very small files (that is, each peer sends you a small chunk of the file). Boosting bandwidth for the first few seconds of a connection can significantly improve the user experience without placing a constant drain on the network.
Unless they advertise it, it's fraud, but it makes sense.
I wish that Verizon did not block ports 25 and 80. I already have FIOSTV and would switch to their internet service in a heartbeat but I don't want to give up my web and mail servers.
The difference between Canada and the USA is that in Canada healthcare is a right and gun ownership is a privilege.
Comcast has been doing this for a while now.
"We are all geniuses when we dream"
- E.M. Cioran
PowerBoost only accelerates the connection if the average speed you've been getting over the past 30 seconds* is less than the speed you are rated at/paid for. So if you have a 6 Mbps connection, that's 768 KB/s max. PowerBoost will raise that to up to 2 MB/s for a little less than 15 seconds, making your average for the past 30 seconds equal to 768 KB/s. After that, no matter how many new connections you open, your connection stays at 768 KB/s. But if your connection gets interrupted/throttled for a few seconds, you may get another boost after it resumes, until you are back to 768 KB/s 30 second average again.
*it may be slightly more/less than a 30 second average. Boosts seem to last about 10-15 seconds, which would make sense with that number.
Iperf, or something like it is what you should be using for speed tests. Set up the daemon on a machine that you know you need to access and tell it to send a ton of data a few times. See what the results are. Those speed tests test how quickly you can communicate with some random server that you'll never need to send any presentations or video files to in day to day business.
I can assure that they do absolutely do this, and it is really annoying.
It's really bad on uploads -- I just ran a test and I got 300 KB/s for the first 5 megs, then it degrades 100 KB/second over the next few megs, so that by the time you have uploaded 14 megs you are getting close to 40 KB/S in upload speed, and the connection is so bad that the shared digital phone line does not have enough bandwidth to have a phone conversation. Stop the upload and start it up again, and you get 330 kb/second, with the same degradation curve.
For downloads they do the same thing, but not so severely -- I downloaded a 67 meg file and it ran at about 750 KB initially, but then dropped to around 350-400 KB/S (according to the FTP app) about halfway through.
So for anyone using the connection for smaller file sizes (like the speed tests) you seem to get "blazing" speeds -- I ran the test at a couple of the internet speed test sites and they both think that I have 12000-14000 kb/s download speed and 2700 kb/s upload speed.
So if I didn't have any other way to measure it, I would think that I was getting way more than I paid for, rather than something that in reality is very pitiful.
I didn't read lots of replies so it may already been stated. What comcast is doing is simple network engeneering. Using a "burst" architeture allow to quickly send small files, at the detriments of long one. Why doing so? simple QoS. It's an easy way to create some degre of QoS without having to introduce more complex technologies. Just think about what are "small" vs "long" files: Small: multimedia using UDP (VoIP,streaming), Web Pages, text emails, instant messenger, most games. In fact mostly what people use their connection for. Long: Email attachments, file downloads, "torrents" Most people can wait for their downloads 2-3 minutes more, but you just can NOT wait a second for a multimedia paquet to arrive. So by doing a "burst" strategy, they give "priority" on small files which doesn't congest your network for long, while slowing the long files that congest everything for quite a time. It's about the same as cars vs buses. By giving more speed/priority to the cars vs the buses, you allow more vehicule to use your road over time. Even if the amount of data transmitted is lower, the amount of different data is higher.
That can't be right. From your description, it sounds like a genuinely good and beneficial to the user idea. Where's the catch ?
Forget magic. Any technology distinguishable from divine power is insufficiently advanced.
Looks like you are getting your 6 Mbps...
768 KB/s * 8 (bits per byte) = 6144, or 6 Mbps.
Since your PowerBoost goes up to 2 MB/s, it would be as follows...
2048 KB/s * 8 = 16384, or 16 Mbps.
Unless there is an problem with the link that can be immediately identified at the time you tested, like a physical problem, then you should develop a baseline of the customers network utilization. Generally, a single download provides insufficient information to in order for to give the employer or customers a recommendation related to their link utilization. This is especially important when the upgrade costs money.
Trending the link utilization is easy to do with free open source tools that will run on Linux, Windows, or a Mac. Or you can pay some $$ and buy software that will perform network utilization trending. Many protocol analyzers have this feature too. As a network administrator/engineer I expect that you can figure how how to tap the link or access the link devices network interface utilization, SNMP, RMON, or even NetFlow/sFlow information.
This is easy to do, just setup an extra PC or laptop on the customers premise for a week, just lock it down logically and physically. Free tools that I regularly use are Ntop (http://www.ntop.org/news.html) and Cacti (http://www.cacti.net). I'm sure that someone on this list can recommend a dozen other solutions.
These tools provide a graphical means to display the link utilization over time, providing greater information over a single download test, thus allowing you to make a more informed recommendation. And the graphics make nice additions to your reports! One scenario would be that the customer is seeing a slow down of their internet connectivity after lunch or late afternoon. Well, trending might reveal that indeed the network utilization peaks at these times when workers get back from lunch and just before they go home. And maybe it's only a few people hogging the bandwidth. On customer networks I've discovered P2P file sharing, large file downloads (movies and ISO's), and even infected computers used as repositories. The customer would have plenty of bandwidth if they just cleaned up that mess or better managed the limited resource with both technical or administrative (policy-based) controls.
And if you have more time, then check out the list of even more network management tools at http://www.slac.stanford.edu/xorg/nmtf/nmtf-tools.html or http://www.ubuntugeek.com/bandwidth-monitoring-tools-for-linux.html.
HTH someone.
-- Wondering how long until the internet becomes fully corporatist, like television.
This makes a lot of sense for trying to make slower cable more appealing to those who want to browse the web primarily. Web sites transfer a bunch and then wait for a while before sending anything else. With a system like this web surfing would be very fast, but doing any kind of downloads or uploads would be much slower. It is a good way to try to charge people more for torrent and high bandwidth usage traffic while offering their ideal market(web traffic) a cheaper option.
AJ Henderson
I've had my Fios Fibre-optic connection for over a year now, and unlike everything else I've had before -- including Crumcast -- Fios has been fast and trouble-free. I can sustain the 5Mb down and the 2Mb up without a hitch, and I've tested this with BitTorrent, of all things.
It's so good, in fact, that it's been exposing problems with my Netgear Wireless Router RangeMax -- I don't think they'd figured on someone sustaining that kind of bandwidth. So it's time for me to upgrade!!!!!!
If you don't have Fios in your area, SCREAM at Verizon. I mean, I've always despised Verizon up till they delivered on Fios. What a rare occasion for a large otherwise stuck-in-the-mud bureauractic company to -- finally -- get it right. And you can't beat the 5/2 service at $40 a month. I was paying twice that for Crumcast!!!!
Fios TV is avaliable in some areas if you still find anything interesting to watch in that medium. For you lucky chums, you'll be able to nuke Crumcast entirely. No idea how good Fios TV is, since I threw out that particle accelerator long ago...
Ruby Neural Evolution of Augmenting Topologies
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.
Some drink at the fountain of knowledge. Others just gargle.
If you want a commercial-grade link you expect to saturate, pay for it! Otherwise, you are stealing from other users and the ISP should throttle you to be fair to them.
How can we trigger this effect all the time?
-- Prepared at the direction of, or to be sent to Legal Counsel, in anticipation of litigation. Attorney Client Pri
any time I run any type of bandwidth testing for clients with Comcast accounts, the results have been amazingly fast
If you want to test sustained speed, then test sustained speed. When you benchmark a HD, do you only test reading small files (which will fit in the HD's cache ram) and then get surprised when the sustained speed is significantly lower? Use the right benchmarking tool, mmkay?
it appears that Comcast is delivering this bandwidth only for a few seconds after any new request and it is immediately throttled down.
And in many situations, especially with consumer links, that's a good thing. Joe User's usage is often bursty (loading a web-page, fetching an email, etc) so designing the network to give good initial burst speed makes perfect sense since it makes interactive usage a lot more responsive.
Is there any valid reason why Comcast would front-load transfers in this way
Better performance for interactive / short-burst content. Which is exactly what many users want.
or is it merely an effort to prevent end-users from being able to assess their bandwidth accurately?
Is it the ISPs fault that you use the wrong tool to benchmark the connection? If you want to know sustained rate, don't use small transfers. One has exactly the same issue with benchmarking HDs or flash-memory, burst and sustained are not the same.
If you can show that Comcast has engineered their network to give priority to traffic from popular network speed test sites, then you'd have a legitimate issue. But from your description it sounds like this is just interactive burst, which is perfectly legitimate and in most situations a good thing.
If J.K.R wrote Windows: Puteulanus fenestra mortalis!
Screw the Romani. They should have stayed on their side of the neutral zone.
None of them can see the clouds; The polished wings don't care.
... This is a so-called "feature" marketed as "Power Boost", "Speed Boost" or "Quickie". Apparently, the expected behavior of this feature is to increase the connection speed of the first few seconds of a download then drop it down to a lower speed. If you want to learn more, there are a few earlier posts that explain this in more detail and dozens that explain it in less detail. Remember - It's only funny the third time.
Never leave a dead horse unbeaten!
Judging just by "feel" though,(& it's the MOST important thing really) I would have to guess you are a Comcast employee pimping your own company, albeit anonymously.
"But this one goes to 11!"
I'm with this guy and I actually like the bursting I get. It allows me to download a file and still browse with reasonable speeds. This is normal behavior.
If you are comparing the speakeasy speedtest to your transfer of the 100MB file the numbers will be different. Speakeasy tests return the results in bits/sec, when you transfer a file the rate is displayed in bytes/sec. So, 15Megabits/sec ~ 1.8Megabytes/sec and 4Megabits/sec ~ .5Megabytes/sec see, both tests were accurate, just displayed in a different format.
So... what exactly stops you from writing some sort of a program that will throttle your network to download 5 mb, wait a second, download another 5 mb, and continue? Your overall average speed will be much higher. It really depends on the "falloff" rate, how quickly you can download again, at the full speed.
If they slow you down after the first 20 seconds into a 700MB iso download, might I suggest:
false; while [ $? != 0 ] ; do wget -c http://ubuntu.com/foo.iso & ; export X=$! ; sleep 20 ; except kill -9 $X ; done
This runs wget in continuation, sets the PID of wget to X, sleeps for 20 seconds, kills wget, and finally quits when wget isn't around any more. Possible problem: if something else gets wget's old PID during the sleep period (after wget finishes normally), kill -9 will kill it, and run through the loop once more.
My ISP was throttling after 5 seconds, preventing me from doing OS updates, so until they fixed their policy, I had to do something similar.
I've been watching this on my company's business cable for a while too. We pay for 6 (maybe 8) mbit Comcast but when I run speed tests I see as much as 24mbit. I also watch the connections on my firewall and Window's Performance Monitor. Why the topic poster thought this was a "speed test defeater" is beyond me. It's obviously a burst feature.
No sig for you. YOU GET NO SIG!
This would explain why Netflix's Watch Instantly feature always stops after a few minutes and has to re-buffer with the message "your internet connection has slowed." It's really irritating. It's been doing ever since Comcast dropped my monthly rate and told me they were signing me up for "faster" service. !@#$%@#$^%
-- QED
The only way to really know is, to install a linux box as a gateway between the users and the ISP and do some serious traffic logging and analysis. This way you can build a profile of how the users are Really loading the system and how the responds. Switching a customer from a low duty cycle ISP that provides bursts of 16 Mb/S to a high duty cycle T1 line that provide 1.54 Mb/S, is a serious endeavor especially considering that the cable modem provides 10 times the bandwidth for 1/6th the price, you need all your duck in a row to make a business case for that.
/mo a T1 is about $300.00 a month.
I'd consider using the Cable for downloads, and bonding a sDSL line or two together for uploads, 1 cable + 2 DSL's is going to run you about $110.00
Apocalypse Cancelled, Sorry, No Ticket Refunds
It's a common practice in networking to allow burst of data at a higher rate than can be sustained for a long period. Years ago frame relay had the concept of a Committed Information Rate and a Burst Size. The customer was promised the full CIR for the long term but could burst several hundred megabytes at wire rate. Today bursting is commonly controlled with a leaky bucket algorithm where a steady flow of tokens represents the allowable steady state rate but you can burst really fast until the bucket empties. The benefit of this approach is that it speeds up interactive traffic compared to long file transfers. If it takes more than a coffee break for something to happen, then taking a little longer won't hurt, but it's really painful when interactive applications slow down. If I had a choice between making speeding up one person's multi-gigabyte movie file transfer or speeding up everyone's interactive work, I'd rather speed up the interactive. Of course, I'd rather have instant everything for zero cost and a million dollar a month salary, but I can't figure out how to get that.
That's for "word of mouth" advertising. It's fairly simple, really. You go to a friend's house, they have comcast tv, and their tv is always on, even if "nothing is on tv". You see and hear comcast ads when you're at your friend's house, and begin to associate comcast with the friend... thus making you more "friendly" towards comcast. Next time you're thinking about being disatisfied with your internet/tv/phone service, you may give comcast a call.
Wow. It's kinda scary that I even think like that. Maybe I should get a job in the advertising sector.
On the other hand, maybe I should just be glad that the only "boob tube" in my house is my 21" CRT monitor. I've been stupefying cable companies for years with the line "I don't have a television" when they offer me some fantastic package deal to go with my internet. You know those "traps" they put on the line so you can get HBO and such? The line coming to my house has one that *blocks* TV signals. Seems they don't trust me not to use their cable to watch television signals on my non-existant TV. On the other hand, it saves me $10/month to not purchase even their "basic" television service, and if they want to install a blocking device for a service I have no intention of using, I'll just chuckle and shrug.
Similarly, I don't get telemarketers, because my house doesn't have a phone. Each person living in it carries a cell phone, as do almost all of my friends who come over. At any given point in time, there are typically at least one, if not *three* phones in my house. So, explain to me again why the house would need its own voice communication line?
--
Yeah, I'm off-topic. So?
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
I have Comcast (unfortunately), and I have noticed that same thing. Whenever I download a large file, like a 700mb Linux ISO image, I will get a super fast download rate for 5 seconds, then it will go down.
I think Comcast uses this to cheat and say "our internet is the fastest ever!". A lot of good it's doing them.
I just heard Verizon DSL is coming to my city in the next 2 months (after I called them up and asked them NOT to send Verizon FiOS ads to me because we can't even get it). The SECOND Verizon is available I am canceling Comcast and getting Verizon
The trouble is modern bloated graphics/flash laden websites, that would load pathetically slowly on dialup...
Because many people think they need broadband, many people get it, resulting in website authors thinking everyone has it. Hopefully the prevalence of mobiles with limited browsers and slow connections will help webmasters see the error of their ways.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
forgot to mention: write a little script named except that reverses the exit code (non-zero to zero, zero to one) for the kill portion. There's apparently a binary on some *nix system somewhere that does that, but I always use a shell script.
In operating system theory, it is well known that a scheduling algorithm called "Shortest Job First" yields the least total waiting time. The SJF algorithm is usually implemented by giving a "new" job high priority, and then reducing the priority gradually as the job accumulates resource usage. The algorithm was developed in the 1960's to allow time-sharing operating systems to provide rapid keystroke response, while continuing to process large batch jobs in the background.
For communication systems, the same principle applies. The only difference is that the network is sharing a different resource (circuit bandwidth), instead of cpu time. The "new" connection gets high priority, and then that priority is reduced as the number of bytes/packets transferred over that connection increases. This allows rapid response for interactive applications, like browsing or editing, while also allowing the network to process large data transfers in the background. To apply it to datagram traffic, the switch just keeps a priority for each source/destination address-pair in cache, and any pair that is not in the cache is regarded as "new".
This has been pretty much standard practice in packet communication switching for a very long time. There is no surprise here, at least not to those of us who have not been doing communications network programming for a few decades.
http://www.dslreports.com/shownews/Slashdot-Keeps-Rediscovering-Comcast-Powerboost-91976
It makes me wonder... If one of the reasons Comcast is moving to switched digital video is to free up bandwidth, when they finally roll it out will we actually reap the rewards?
"Women. Can't live with 'em. Pass the beer nuts." -Norm
I have had comcast for quite some time now. They introduced "speedboost" initially on the top tier plan but have extended it to the lower plans as well. It operates as a token bucket throttle, it may in fact be a token bucket, but I do not know how they are accomplishing it. So you start with a bucket of tokens, you are allowed to expend the tokens as fast as you can up to about 24Mbps, once the bucket is empty you are out of BW. The bucket refills at a rate equal to your plan rate.
I run similar filters in the linux firewalls I set up for clients as well as general BW throttle and fairness queuing. Anyone interested in this stuff for their home/bus network should look at the tc filters available in any linux kernel.
Anyway, people can bash comcast all they want about various crap they pull, but this particular feature works very well for normal user type connections. Very high initial burst rates allowing the first 10-20 MB to come in very fast then the throttle kicks in. Great for browsing and smallish downloads.
I have a ten meg connection at my house but I download files from a server that has a 1.5 MBPS T1. The server I download from or upload to may have 15-30 others doing the same thing. I have no idea if Cox Cable is keeping me from my 10 megs because I have a Ferrari that I am driving on a one lane road behind a street sweeper. The internet industry has based its model around T1's so what can we do other than boast we have a high speed connection and hope the others have it too. What are the chances the network path chosen by the routers are all large pipes.
When it comes to testing my connection, who is to say that my ISP does not have a high Quality Of Service applied to the most popular bandwidth testing sites, that would be good for business if QOS was enabled for those test sites.
By the way,,, I have my entire house wired with CAT6, gigabit switches and gigabit NIC cards so I can pummel the crap out of my 10 MBPS Cable modem, I am thinking about changing to fiber, your thoughts!!
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
I believe new wave communications is also doing this.http://www.newwavecom.com/ Wondering if speakeasy would be better or not. A little more extra dollar but what is the value comparison ?
Cable allocates TV channels for data.
You can be allocated a whole channel to yourself, or just timeslots on a shared channel.
Channels are allocated on demand. If you don't use your channel, it is given to some other user. When you start to need it again, a channel is given to you. There is some latency in granting a channel, and some latency in waiting for a timeslot on a shared channel.
As long as you have a dedicated channel, your connection is fast and low-latency.
This is not at all uncommon. Having worked as a network engineer for an ISP for 7 years just prior to my current job, I can tell you that this is common practice, especially if they are using any ATM. One of the reasons is that most session based transactions (web pages, email downloads, etc) are over in a few seconds. It actually provides better throughput and congestion control for the entire network to allow the initial transaction to burst at a higher speed, since a huge portion of those transactions are over very quickly.
:)
Think of the connection as a large pipe (your cable connection) with a small outflow valve (your modem), connected to a larger, higher pressure pipe (your ISP). Until your local pipe is full, you can put water into it as fast as you desire. But once it is full, the volume slows down because you can only put in as much as you are taking out (your cable modem connection/outflow valve). So what speakeasy and various other speed testing sites see is the effect of filling up your local pipe (your connection to your ISP).
What a large file download shows you is the actual throughput.
BTW, this is also a quick, very simplified explanation of bandwidth (how much data you can pack into the pipe) vs. throughput (how fast you can actually pull data through the pipe).
Comcast calls it PowerBoost - basically, it's a marketing tactic. It starts at 15mbps down and throttles back after about 15 seconds or so on the assumption that most users only download ~10mb at a time for a "large" download. They actually call it a feature and use it in their advertising (targeted at idiots and users who don't know better).
#include <disclaimer.h>
#include <beer.h>
Airwave hosts small networks at residential complexes (apartments, condos, etc). They have very similar behavior to what you're describing about Comcast. Downloads will go very fast unless they go on for an extended period. Browsing and other small traffic is incredibly fast.
Although the feature is very handy so that browsing traffic and other light average bandwidth applications get high speed, I also noticed I can do speed tests up to about 760kbps down and 200kbps up. Never in the last six months of using this connection for any application whatsoever have I observed speeds that fast to any server at any time for any duration. This includes transferring files to a server on campus nearby. Not for the first second even. To test to see if they were simply allowing speed test traffic to be un-throttled, I left a download going for an extended time (as usual speed decayed over time to a minimum of about 15kbps) and then opened a speed test. >1MBps up 200kbps down. My conclusion from repeating this several times is that Airwave is in fact giving bias to speed test sites in order to distort users' perceptions of their connection capabilities. Airwave isn't a large operation, but it goes to show that this kind of behavior is happening.
I should mention that Sprint is the ISP who hosts the T1 line(s) connecting our network in the complex to the internet.
"There are some people that if they don't know, you can't tell them." ~ Louis Armstrong
After reading and commenting on Slashdot for some time, I stopped reading it months ago due to dumb stories like this. I only came over here because Broadband Reports was making fun of Slashdot:
http://www.dslreports.com/shownews/Slashdot-Keeps-Rediscovering-Comcast-Powerboost-91976
kdawson continually posts garbage like this. s/he clearly has no clue. Get rid of kdawson and maybe I'll come back here.
Penny - plain text accounting
I know it is popular to bash Comcast these days, but I am a Comcast subscriber and I regularly get about 1 MB/s download speed, when downloading actual stuff, not benchmarks. One recent example was the Sun Java SDK. This is under Linux BTW, although I don't think it matters. So, I am pretty sure that they are not cheating, at least in my case.
... but the biggest one is that file size (like a lot of other things) has a "long ail" distribution, ie, if you had a histogram of file asizes,there would be many more of them than big files. So if you weight bandwidth to maximum speed for little files, you dramatically improve average throughput.
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.
Innocent people shouldn't be forced to pay for inferior software development.
--"Code Complete" Microsoft Press
Never, cant be true.
---- Booth was a patriot ----
Lots of interesting explanations here..
This could also equally be normal, every-day traffic shaping.
Even simple shaping algorithms let you set some level of overflow before you start dropping packets.. it smooths things out.
My cable company gives me a PowerBoost feature for free. Your contract promises you a minimum amount of bandwidth. When the cable company has extra bandwidth available you get more speed for free.
I've noticed that even when a downloading DVD iso of Linux (gigabytes in size) that the speedboost kicks in pretty well. Some days I get more and some days I get less. I've always had my minimum promised speed though.
This mirrors my Comcast experience exactly. I've noticed that the various online bandwidth testers invariably show a much greater link speed than achieved with large downloads. Typically I can get about 750 KBps downloading, but much higher bursts in the beginning of the download. The link will start at about 1.5 MBps then scale back consistently to about half that. Basically, these guys are stealing my money if you go by the advertised speeds. Still, it's faster than my old DSL line. Verizon is busy stringing up FIOS in the area and if it does turn out to be a better deal then Comcast can kiss my butt.
I have mod points. The reign of terror begins now.
is that ISPs have always done this. In effect, this behavior makes things like HTTP and small file downloads seem very snappy at the expense of larger file transfers.
It actually isn't that bad of a policy in that it prioritizes real time operations over batch jobs. Your CPU scheduler acts similarly. It does become a problem if the base rate really sucks.
My adblock plus blacklist is pretty short. Not all that many adservers really treat you like prey, and adblock lets you treat them the same way in return. It's a shame so many of /.'s ads are animated. Every time I unblock them I stop feeling guilty about blocking them again.
As always, all IMO. Insert "I think" everywhere grammatically possible.
The article summary sounds for all the world like a description of a burstable connection, which is widely considered a selling point. I'm not sure how _valuable_ a feature it is, but a lot of ISPs advertise it, along the lines of "768 KBPS downstream bandwidth, burstable to 1.5MBPS" or similar. This has been around for a long time. In fact, I am almost certain burstable connections were around before DSL.
Cut that out, or I will ship you to Norilsk in a box.
I tend to browse in frames with about 5 to 10 frames open at a time, and a separate window per topic I am browsing. Let's say...one for comics, one for Slashdot and one for whatever I am really doing at the time. My ISP sucks for latency when I am doing things like games, but if I am doing lots of small DL or some large-ish DL (larger than 1228x1024 images) then I get good performance this way and when I am done with each page, the next is usually ready. I have not downloaded any iso's lately to tell you how that goes.
Phil
Laugh, it's good for you!
Anyone sending or receiving attachments via email will be happy for Power Boost. It will download at very fast speeds the first 25MB of a download. Clears the pipes so to speak.
I can understand not showing the ads to people in comcast service areas, as they wouldn't be potential customers anyway...
But only showing the ads to existing customers? How do they intend to attract new customers?
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
>>>>> Broadband, like dialup, is subsidized by the low use casual customers.
>>> In other words the casual user is paying too much.
Well...
That's true, but only because the "casual user" didn't do their research. People can still get $7.00 a month Netscape dialup accounts (unlimited usage). This is what I have for travel, and it works just fine for web-browsing. And there is a wide variety of other plans available:
- free
- $7 for dialup
- $15 for minimal broadband
-~$40 for standard broadband
-~$80 for maximal broadband
The various companies offer the users different options based upon how much bandwidth they think they will need. If a low-usage person pays more than what he/she needs, well then maybe they should downgrade & save some money. (IMHO).
Conclusion:
Per usual, it comes down to the casual, low-usage customer cheating him or herself. If they were wise, they'd get a cheap $7 dialup or $15 DSL connection, but because they didn't do their research, they end-up paying around $40 a month --- far far more than they need to pay.
And that's their problem, not mine.
I pay my fair share ($80) for my habit.
The government is not your daddy. Its purpose is not to raid middle-class neighbors' wallets and give it to you.
A T1 may only be 1.5mb, but its 1.5mb both ways.
Also you're far more likely to get other useful services with a T1 like multiple IPs, reverse DNS, an SLA offering guaranteed uptime, and the ability to use the full 1.5mb in both directions day in day out with no fear of disconnection, contention or shaping. Consumer ISPs will usually slow down (due to shaping or congestion) during busy periods, and they might terminate your connection if you use obscene levels of bandwidth. With a T1, you are paying for the 1.5mb dedicated to you, not shared with anyone else.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
Think of this as education. /. for objective information don't understand the difference between bandwidth and throughput. Unfortunately I could not find any end user suitable tools for measuring throughput. While Stanford has a decent list of tools it's by no means exhaustive and they're not packaged for the end user.
/. and immediately called your legislature to complain about it? Now if you'll excuse me, I have to compose an email to WHO and threaten to withhold my pittance of support of they don't get their $#|t together.
While it's common practice, most consumers who look to
http://www.slac.stanford.edu/xorg/nmtf/nmtf-tools.html#thruput
FWIW, the poster's measurement of throughput, crude though it is, is way better than I see through Time Warner during normal business hours. It's unfortunate what we yanks pass off as Broadband, but then again; when was the last time you read about a horribly obtuse piece of pork barrel legislation on
Yes, you are correct on all points, my frustration is with the belief that a T1 is a lot of things it is not. Most people believe 100mbit Ethernet is slower than a T1, that it is somehow this amazingly fast ultimate god of connections. In today's market, a T1 is a slow, over priced relic of a by-gone era. For the same money, you could order an ethernet loop at 5mbit up and down. You get the guaranteed bandwidth, the reliability of a commercial circuit, the 100% balls to the wall control over how the bandwidth is shaped, all the stuff you just said, probably save a few bucks a month in the end, or at least break even.
Unless you live in some backwater little berg, you should almost never be caught dead ordering a T1 loop. Modern technology can deliver more capacity and reliability, at a lower cost.
--Nuintari
slashdot : where an opinion can be wrong.
This agrees with my observations here too. I guess it's a feature, because more bandwidth is always a feature.
I regularly download Firefox and Thunderbird updates with speeds around 1.4MB per second. It is about 12Mbps. So not just test sites are fast. - Alexey.
It's not just IMDb.com. Vodafone have a customer accounts webpage that allows anyone to check the balance on their account, which is great for wireless modems. Except of course, you have go through three pages of corporate logos, login prompts and adverts.
Going back to the days of dial-up speeds really makes it obvious where all the bandwidth is going... fancy full-screen wide curvy squiggly frames, flash movies, clicky-soundy buttons, pop-up animations, fancy animated sparkly animations that follow the pointer around...
Thank goodness for text only browsers.
Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads