throttling web performance : performance testing

Have you ever wondered why your web access speed is sometimes blazing fast, while at other times you are waiting for response at a slow-as-molasses pace? Yeah, me too. I’ve been obsessed with web performance and page speed since 1996, so I’m sensitive to slowdowns.

Would you be surprised to find out that telecom/cable providers are intentionally and deliberately slowing down web access? Yeah, me too. Empirical data from test results show that at least one provider is chocking users access 85% of the time! U.S. providers are slowing you down twenty-three percent (23%) of the time. Globally, it’s even worse – 32% of the tests show provider slowdown.

There is software that proves it. Internet access operators don’t want to talk about it openly, but when you dig into their standard service contracts, it’s confirmed. They “throttle” your speed whenever they want based on their needs…not yours.

NY Times Sheds Light on Throttling Slowdown Sources

Data speed was critical to me in 1991 because I wrote proprietary healthcare information exchange technology back then using 2400 baud modems. The kicker? The programming language was Clipper & Foxpro. Sounds stupid to me too…looking back. But it was fun programming, a bit cutting edge, and I got ego boosts whenever my geeky friends would ask, “You did WHAT?! With a room of 286 PCs and a Clipper?!”

It was a glorious day for me and my rag-tag team of basement-dwelling coders when we finally received a 9600 baud modem in the UPS shipment. We had a party that day!

Twenty years later, and I am still ultra-sensitive to data transfer slowdowns. I’m usually very happy with my Comcast connection, but occasionally my web applications are slowing to a crawl. It’s obvious from my app server monitoring that the problem is somewhere other than the back-end. Thanks to this article published today by Kevin O’Brien at the NYTimes.com I have a better idea of why my Internet access speed seems to fluctuate inexplicably.

Kevin shares with us how the Networked Systems Research Group concluded:

…the blame often lies with the telecom operator, which is selectively slowing broadband speeds to keep traffic flowing on its network, using a sorting technique called throttling.

Based on data gathered by Glasnost, a free software gauge that detected whether broadband service was being throttled by a network operator, shows that “…throttling is being done everywhere in the world.” Glasnost was released in 2008 and was developed by Krishna Gummadi & Marcel Dischinger at NSRG. The software has been used globally by 1.5 million people to capture measurements of access speeds to detect throttling.

32% Is The Global Average of Network Throttling

At least here in the United States, throttling isn’t as bad as other places in the world. Our Internet access providers are more merciful than their colleagues in Britain.

In the US throttling was detected in 23 percent of tests on telecom and cable-television broadband networks, less than the global average of 32 percent. The U.S. operators with higher levels of detected throttling included Insight Communications, a cable-television operator in New York, Kentucky, Indiana and Ohio, where throttling was detected in 38 percent of tests; and Clearwire Communications, where throttling was detected in 35 percent of the tests.

I guess I’m feeling better about living in Colorado and having Comcast; at least we weren’t named as especially bad for intentionally choking our speed.

What others were mentioned? Here is a brief summary of larger operators around the world and the percentage of tests that detected throttling:

  • Rogers Communications (Canada) – 85%
  • British Telecom – 74%
  • Bell Canada – 64%
  • Opal Telecom – 50%
  • Telewest Broadband – 50%
  • NTL – 50%
  • Tiscali U.K. – 50%
  • Pipex – 50%
  • NTT Docomo – 49%
  • Kabel Deutschland – 44%
  • Vectant – 38%
  • Insight Communications – 38%
  • Clearwire Communications – 35%
  • AT&T WorldNet – 30%
  • GigaInfra Broadband – 30%
  • France Télécom – 21%
  • Numericable – 21%
  • Neuf Cegetel – 21%
  • AT&T SBIS-AS – 18%
  • Verizon landline – 18%
  • Deutsche Telekom – 16%

Wide Disparity Between Other Operators

Smaller Internet access providers have a disadvantage in the marketplace when it comes to building out broadband networks because they typically have economy of scale issues that lead to lower profitability; thus, they have a far harder time justifying the capital investments in creating higher capacity infrastructure. That of course has resulted in their increased usage of throttling.

Here are some of the small operators mentioned and the percentage of tests detecting throttling:

  • Emirates Integrated Telecommunications (Dubai) – 90%
  • Toya (Poland) – 88%
  • Tele Centro (Argentina) – 87%
  • RLE Elisa (Estonia) – 85%
  • ASN AtHome (Hong Kong) – 83%
  • TM Net (Malaysia) – 78%
  • Magix (Singapore) – 63%
  • Cabo TVM (Portugal) – 62%
  • Bezeq (Israel) – 59%
  • Telstra (Australia) – 34%
  • Telefónica of Spain – 19%
  • Telecom Italia – 19%
  • KPN (Netherlands) – 18%
  • Eircom (Ireland) – 15%
  • Telia (Sweden) – 14%
  • Belgacom (Belgium) – 13%

Why Do They Throttle Us? Will It Continue?

O’Brien believes that access providers aren’t using throttling as an impediment to web performance; rather, it is a key mechanism to keep “bandwidth hogs” from consuming more than their fair share of the Internet access pipeline.

The above percentages showing how broadband operators are for landline providers. Mr. Dischinger acknowledges that mobile networks are much more prone to use throttling. None of those statistics were available in the Times article. I’ll try to research that tomorrow and provide more information relative to Mr. Dischinger’s assertion. It certainly makes sense, so I believe he is absolutely right. I would suspect that mobile networks are throttling in 100% of Glasnost testing.

Mr Dischinger also believes throttling will increase because the need for controlling consumption will be greater as landline services continue to spread to more homes where users will need more and more bandwidth. I agree. In fact, I’m probably one of those bandwidth hogs. I love watching my daughter’s rugby games streamed, and I’m a regular on ESPN3.com to watch Kentucky basketball games that aren’t available here in Colorado.

I highly doubt it can go on forever. I cannot envision with the current network infrastructure they have that operators can continue to support people in the long term without more investment.

Amen, Mr. Dischinger! Thanks for your eye-opening research and very cool tool that uncovers the choking ways of providers.

Similar Posts