The Computer Corner

Internet data: How it gets from here to there
By Charles Miller, March 2, 2007

The subject of a recent column was “Network Neutrality” and its importance to the internet. This week I want to fill in some of the gaps in the earlier column.

A term you will be hearing more in the network neutrality debate is “QoS,” short for “quality of service.” This is a networking term that specifies a target throughput level at which end-to-end latency is within the desired range. In plain language, this means your internet communications get through from source to destination before it is too late.

All internet traffic is broken up into data packets. The email you write and the photo you attach to it might be broken up into fifty packets which are sent from your computer, then the fifty packets are reassembled by the receiving computer.

Decades ago the designers of the TCP/IP communication standards used on the internet designed in some incredibly far-sighted capabilities. Among these were four “type of service” bits and three “precedence” bits provided in each packet. The early internet lacked the ability to use these to provide QoS because its infrastructure was too limited, so this information was ignored though it was there. Today QoS data are being used more and more frequently, and this brings into question how it is, and should be, utilized.

Many things can happen to packets as they travel from origin to destination. Packets can be dropped (lost) or delayed. They can be received out of order or reach the destination with different delays (jitter).

Different packets may take different routes, each resulting in varying delays. The result is that the packets arrive in a different order than the one in which they were sent. This necessitates rearranging out-of-order packets to an isochronous state once they reach their destination.

For simple applications such as email, this is not difficult but for streaming multimedia (audio or video) and Voice over IP (VoIP) the quality is dramatically impacted if the packets fail to reach their destination in time.

Video teleconferencing (VTC) and VoIP both require strict limits on jitter and delay. Monitoring and signaling applications (burglar alarms) are termed “inelastic,” meaning that they require a certain QoS. Delays in audio or video are immediately noticeable as would be a false triggering of a burglar alarm.

Elastic applications can take advantage of however much or little bandwidth is available. If your email program takes an extra ten seconds to assemble an email, you would never notice.

Beyond the technical considerations are the practical. A medical doctor doing remote surgery via VTC is understandably going to claim his bandwidth is a lot more important than teenagers engaged in idle banter in online video chatrooms.

The internet relies on congestion avoidance protocols built in to TCP/IP to reduce traffic load whenever the system is overloaded. Applications such as VoIP and VTC require constant connectivity and cannot reduce their traffic rate to help prevent an internet meltdown. Other applications such as email, file downloads and such are elastic and can easily throttle down their bandwidth consumption to half or less to reduce the traffic load on the system.

The question of how much priority should be given to each type of internet traffic goes to the heart of the network neutrality issue.

Charles Miller is a freelance computer consultant, a frequent visitor to San Miguel since 1981 and now practically a full-time resident. He may be contacted at 044-415-101-8528 or email FAQ@SMAguru.com.