Small device... important data
In one of my recent posts I wrote about the need for fast broadband connections in Europe. We need more and better broadband access to deliver the digital agenda and get every European Digital. And we need to know more about the current state of the market – so we’re looking for volunteers across Europe!
Over recent years, we have made good progress in terms of broadband take-up; advertised average speeds have also gone up. But what about “real” broadband speeds? Many times I hear people say they upgraded their broadband connection to a “faster” product - but then can’t notice any difference between one and another. But then, what’s the point in paying a more expensive monthly fee?
The disappointment is often caused by a difference between advertised speeds, and what you actually get for your money. You’ve probably seen speeds of broadband products advertised as “up to” X Megabits per second. This speed is called “advertised”, “nominal” or “headline” speed and it is what we see in the ads. In reality, a broadband connection can never achieve 100% of the advertised speed, because part of the bandwidth is reserved for IP addressing and routing. However, it seems that, sometimes, the effective speed is significantly lower than what we could expect.
To trust in the market, people need to be sure what they’re getting. If not, if European consumers are unhappy about their broadband access, we won’t succeed in building the Digital Single Market, because people won’t have the confidence to sign up and pay the price for the high speed connections needed for new and innovative services.
I want us to understand this problem: I want us to collect first hand information from European consumers on the actual quality of their broadband access. To do that, though, first of all we will need accurate, comparable and reliable data on effective speeds at EU level.
You can help me collect this data! We are looking for volunteers across all EU27 countries plus Croatia, Iceland and Norway. If selected as part of the “representative sample”, you will be sent a small device to plug into your home internet connection. This will run a series of automated tests when the line is not in use, establishing the speed and performance of your broadband connection – helping us to get the data we need to get every European Digital.
If you’re interested in being one of this community of volunteers helping – visit the site now!
Number of views: 6971
Tags: broadband, broadband speed, Digital Agenda, fast broadband, samknows
Email This Post
RSS comments feed on this entry
> In reality, a broadband connection can never achieve 100% of the advertised speed, because part of the bandwidth is reserved for IP addressing and routing.
It’s more that, for DSL, the de-facto speed is directly proportional to the line length of the last mile of copper. A customer in the direct vicinity of the building housing the DSLAMs might get the advertised speeds, but there’s a steep falloff with distance. I live 2.8km from the DSLAM and get 10 instead of the advertised 16MBit (and the former monopoly competitor ISP would even only guarantee 3MBit to me). The ISPs have differing standards.
Where one provider would disregard possible transmission losses and future interruptions to be able to boast with high speeds in the market, the other errs on the side of caution and only offers low speeds.
Besides, what is the budget for this data acquisition project?
Please open up the cable systems in the netherlands for other competitors then Ziggo and Upc. This will give new companies access until now monopolised infrastructure.
The transport medium also adds overhead, which may be nonlinear. Take for example ADSL and ADSL2+. These connection types encapsulate the IP traffic in ATM cells. Each ATM cell is 53 byte in size, and carries 1-48 byte data. That means that a packet, which looking at the IP layer is 49 byte in size, requires two ATM cells at the ADSL transport layer, so it actually uses 106 byte of connection bandwidth to transfer a 49 byte packet. In other words: It’s not sufficient measuring the bandwidth using large packet sizes, because the bandwidth may depend on the sizes of the packets being transferred.
If the overhead was linear, it would be simple defining requirements for the ISPs: Advertise the net bandwidth, not the gross bandwidth. We don’t buy meat, potatoes, flour etc. by weight including packaging, so why should we buy bandwidth by bitrate including packaging.
However, since the overhead is often nonlinear, the bandwidth could be advertised as an interval, showing worst case and best case numbers.
Recent packet size statistics shows that most users on average download relatively large packets (1000 byte/packet) and upload relatively small packets (less than 200 byte/packet).
Additionally, advertising average bandwidth is not sufficient. I don’t want my 10 Mbit/s connection delivered as 100 Mbit/s in 1 second followed by 9 seconds of 0 bit/s.
In fact, most users don’t care about bandwidth; they care about response time, where bandwidth is only one parameter. Latency (“ping time”) should also be advertised.
At SmartShare Systems (http://www.smartsharesystems.com) we do have ISP customers that use our bandwidth management appliances to optimize bandwidth for their subscribers. Our appliances show exactly how much bandwidth is available for each subscriber in real-time (i.e. every second, not just 5 minute averages) and what the connection latency is. This ensures that their subscribers get the internet experience they want, not just some stupid bandwidth number you can stick on a Service Level Agreement.
> It’s my dream to get Every European Digital
With your most recent acquisition (the latter to be taken literally) of the lying baron zu Guttenberg, European citizens now are stuck between a rock and a hard place. Hiring that prominent promoter of an almighty police state and his credo of “security at any cost” was the stupidest thing you could do! Is this what you dream of?! Ensuring every European gets online in order to be entirely monitored, Orwellian style? Thanks to INDECT, ACTA and numerous further foulnesses to come? Zu Guttenberg is already the second dangerous “EU talent” along with Mrs. Malmström…
why not utilise http://www.google.com/publicdata/explore?ds=z8ii06k9csels2_&ctype=l&strail=false&nselm=h&met_y=avg_download_speed&scale_y=lin&ind_y=false&rdim=country&tdim=true&hl=en&dl=en&iconSize=0.5&uniSize=0.035#ctype=l&strail=false&bcs=d&nselm=h&met_y=avg_upload_speed&scale_y=lin&ind_y=false&rdim=country&idim=country:GB:SE:NO:DK:NL:DE:FR:FI:LV:LT&ifdim=country&tdim=true&hl=en&dl=en
That service provides upload and download speed data and is based on billions of measurements from speedtest/Ookla etc.
The Ookla/speedtest results depend on the user structure, the user hardware and the time users test, the samknows approach is more objective, with hardware plugged in the wall, before the user’s computer gets involved.
Will be interesting to see how the results of the two approaches compare.