Mobile bandwidth problems are telco's fault

WHILE TELECOMM COMPANIES complain that smartphones are bollixing up their networks, a blogger thinks that the problems might be of their own making.

Writing in his blog, Brough Turner has come up with good evidence that AT&T's problems with the Iphone might have nothing to do with having too many fanboys on its network at once.

He said that at least some of the telco's problems are configuration errors. He claims that congestion collapses are being created by misconfigured buffers in its mobile core network.

The technical data behind Turner's thinking is based on a painstaking reading of the IRTF's "end-to-end" email list.

He found a post from David Reed, who noticed terrible throughput and extreme delays from overly large buffers in the routers and switches in AT&T's core network.

Outfits that use multiple 3G data links on the Verizon Wireless, AT&T Wireless and Sprint PCS networks have noticed the same thing, but mostly just on the AT&T Wireless network.

Telcos seem to have set their routers to use excessive buffering in order to absorb transient traffic bursts. This is barking, as research and operational experience show that there is no simple formula for how much buffering is optimal at any given location in a network.

To get around this problem there are some good basic precepts and it appears AT&T Wireless is violating one of the most basic ones.

The buffer in front of a congested link must induce some packet loss. TCP increases its transmit rate until it experiences packet loss, then it cuts its rate in half and enters a congestion avoidance mode.

"It appears AT&T Wireless has configured [its] RNC buffers so there is no packet loss, i.e., with buffers capable of holding more than ten seconds of data. Zero packet loss may sound impressive to a telephone guy, but it causes TCP congestion collapse and thus doesn't work for the mobile Internet!" he wrote.

Oops. µ