Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

On Windows/Linux, what would be a way to run a sample measurement of the bandwidth that I have available? I need to run a measurement at periodic intervals. There are many website which will give you a measurement if you load the page in a browser. But I want to take many data points over several weeks, and these websites do not offer the data as a service, but as a flash UI, which is very hard to scrap. Also, Ideally I would like to make benchmarks using specific sites and url queries, and these tests are a bit obscure about what they are truly measuring

My problem is that It has become pretty obvious to me that my ISP is downthrottling the bandwidth of my home connection during business hours, assuming no one is at home, or that someone at home will not use much bandwidth. Since last time I've checked, my contract does not specify anything like this, I'm hoping to gather enough data over a period of a month to present a formal complaint about the bandwidth statistics and how they correlate with hours where the service seems to be down throttled.

share|improve this question
Have you had a look at ubuntugeek's Bandwidth monitoring tools page?for the linux monitoring also the "downthrottling" is probably just congestion on your line unless your using a 1 to 1 Contention ratio – 50-3 Aug 12 '13 at 23:00
Spiceworks has this feature, but might be overkill for your task. – Moses Aug 13 '13 at 16:04

There is no 100% accurate way to do this since your speed depends on many factors. You can do some kind of periodic speedtest to measure speed at some point in time, and make crons to run it when you want.

First you need to know that the server you are downloading data from can actually send it at the speed greater (or equal) then the speed you expect (you pay for). Second you need to know that there is no congested link between your ISP and the server, or it will affect your test. You cannot know this, but picking a server that is physically closer to you is preferred. Since you are paying for access speed to your ISPs network, they can not be responsible for congestion happening somewhere else on the internet. Also if you don't have some monitoring software that will show all bandwidth used by all hosts on your network, you should have no other activity on your link beside the test data transfer at that time. You can download any file, i usually download some linux .iso from a server that is close to my location. Monitoring software can display your used throughput at some time but this is not necessarily your maximum available throughput at that time (for the reasons mentioned above).

To be honest i don't see a very good reason for you to do this, it will render your connection virtually unusable at the test periods and i don't think you will get anywhere with your ISP. This looks like simple congestion and not something that is deliberate, and they will probably tell you that they cannot guarantee the maximum bandwidth (unless you have that kind of contract). Also, if they assume no one will use much bandwidth at work hours, i don't see what would they gain by capping it.

share|improve this answer
"if they assume no one will use much bandwidth at work hours, i don't see what would they gain by capping it" - well, maybe they think there is not enough people willing to demand good service at those hours, simply because statistically the people that complain is a smaller fraction of the people that actually suffers the problem – lurscher Aug 15 '13 at 4:22
Yes of course, but I meant what would they technically achieve by doing that. Traffic shaping is done to avoid congestion, if there is less users using bandwidth at that period there is no apparent reason to shape traffic since it is less likely the link will be congested. – pajaja Aug 15 '13 at 7:55
Btw if they are really doing this, another indicator would be constant download speed. For example, if you pay for 10mbps and they are limiting to 6mbps, when you are downloading large file the transfer speed would be constantly 6mbps, without significant changes. If the speed varies I would say it is a problem with lack of bandwidth for all users or something else (depending on what type of connection you have). – pajaja Aug 15 '13 at 7:55
pajaja, what I believe is that the benefit is to allocate the existing bandwidth to business accounts, instead of residential ones (that pay higher rates) but avoiding larger investments in infrastructure – lurscher Aug 16 '13 at 1:34

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .