Sam shines a light on BT's traffic shaping

 

Gallery

By Barry Collins

Posted on 4 Aug 2008 at 09:02

Enlightening new broadband performance tests reveal the scale of traffic shaping taking place on BT's broadband network.

The Sam Knows Broadband website has been conducting a six-week performance test of British broadband, by issuing more than 200 volunteers with specially adapted routers that record the speed and other metrics associated with their broadband connection.

The tests showed that BT's broadband service provided the fastest throughput when measured as a percentage of what the site calls "implied line speed" - the maximum potential speed of the connection.

However, Sam's research reveals exactly why BT is performing so well: the controversial practice of traffic shaping.

Whilst speeds on Port 80 - the one used for regular HTTP traffic - were very impressive for both BT and its subsidiary ISP, PlusNet, traffic on the ports typically used by peer-to-peer applications dropped to around just 15% of the line speed in peak hours.

"The cause of this can without doubt be attributed to traffic shaping - the practice of prioritising one type of traffic over another," the Sam Knows report claims.

"PlusNet openly admits and advocates traffic shaping policies on its website, so these results were to be expected," the report continues. "However, BT is not so forthcoming, and the breadth and scale of the activity is rather surprising. In fact, the same characteristics can be seen when looking at any of the BT connections we monitored individually - including business connections and also including two connections that were used only by the monitoring units we installed.

"This suggests that the policy is applied universally, regardless of product and regardless of usage volume."

Speaking to PC Pro, the report's author Sam Crawford, said the traffic shaping wasn't necessarily a bad thing. "On BT, we're seeing HTTP traffic running at very high speeds, but other P2P port ranges see massive dips at peak hours," he says. "This traffic shaping helps its HTTP tests to perform so well."

Virgin variations

The report also reveals a wide variation in performance from Virgin Media's cable services. Whilst the actual throughput of Virgin's lower-speed products almost matched their headline speeds, the actual speeds of its top-of-the-range 20Mb/sec product was highly variable.

"The 2Mb/sec and 4Mb/sec products maintain their full speed consistently, and the 10Mbps product does a good job too, apart from the occasional dip," the report claims. "The 20Mb/sec product is where the issue lays, with significant dips seen every evening."

"Significant variation was seen between individual 20Mb/sec lines, with many running at 20Mb/sec during off-peak hours, but some dipped down to below 10Mb/sec."

Sam Crawford says Virgin's performance varied from area to area. "The 20Mb/sec product doesn't come close to 20Mb/sec on average," he says - it actually recorded an average of 14Mb/sec. "It dives dramatically in the evenings."

"There are regional capacity issues," he adds. "Some areas have better capacity than others."

Crawford also has doubts over Virgin's ADSL service, used in those areas where it can reach customers with cable. "Virgin's ADSL service is pretty poor," he claims, with the ADSL service coming bottom of both Port 80 and non-Port 80 speed tests.

"It runs over BT Wholesale and it [Virgin] hasn't bought sufficient backhaul capacity, I imagine," he claims. "If you don't buy enough backhaul you're going to see these kinds of problems. It's down to core capacity."