The Time Square Ball bringing in 2008 had more than 9,500 LED bulbs displaying 16 million colours while consuming power equivalent to about ten toasters. This compares to 600 incandescent and halogen bulbs adorning last year's Ball.



Easy to forget that most mobile devices used by Time Square revelers were behind IPv4 NAT's and that always on applications such as Instant Messaging, Push e-mail, VoIP or location based services tend to be electricity guzzlers. It so happens that applications that we want always to be reachable have to keep sending periodic keepalive messages to keep the NAT state active. Why is that so? The NAT has an inactivity timer whereby, if no data is sent from your mobile for a certain time interval, the public port will be assigned to another device.

You cannot blame the NAT for this inconvenience, after all, its role in live is to redistribute the same public addresses over and over; if it detects you stopped using the connection for a little while, too bad, you lose the routable address and it goes to someone else. And when a next burst of data communication comes, guess what? It doesn't find you anymore. Just think of a situation we would loose our cell phone number every time it is not in use and get a new one reassigned each time.

Luckily there's a way around but it requires a constant stream of keepalive message sent at preset intervals, often 30 second or so. These dummy messages give the NAT the illusion that the communication is still on and it doesn't disturb your address. Only drawback is that this happens to drain the battery and shorten standby time considerably. Just try leaving IM off instead of on to see the difference.

How much could this possibly drain the battery? In their excellent study [PDF], Nokia's Haverinen, Siren and Eronen tackled the issue. In a 3G environment, 20 second interval keepalive messages result in an average current of 34mA and this goes down to 6.1mA if no keepalive messages sent. If we know consider a Lithium Ion battery which is 3.6V and remember that Volt times Ampere gives Watts, this looks like 100.44 milliwatt saved. Imagine 3 billion 3G phones in 5 years time or so and bingo, roughly 300 megawatt of useless keepalive message. If we assume that all 3 billion 3G phones would be in use two hours a day on average, this still leaves 275 megawatt wasted.

How many power plants does it take to generate 300megawatt?

How many tons of carbon dioxide in the atmosphere to produce 300 megawatt?

An article in EE Times Europe estimated that mobile networks consume 61 billion kwh and that this could double by 2011. Do cell phones in standby draw less power from the base station than if they communicate all he time? A higher density of phones per cell site conceivable if inactive phones in standby instead of a quasi permanent state of excitement? Who knows, this could maybe save some additional megawatts, this time by the operator.

And while on the power topic, all of us running major networks or content server farms take note of their electricity bills. Jonathan Koomey of National Livermore Laboratories estimates in his study [PDF] that servers worldwide consumed a cool 123 billion kwh in 2005 for a total electricity bill of 7.2 billion dollar!

Did anyone recently voice the opinion that IPv6 and energy saving was just marketing hype? The Time Square Ball LED light bulb initiative sure must have had its detractors too.

Happy New Year!

Any opinions, findings, conclusions or recommendations expressed in these articles are solely those of the author and are not in any way attributable to nor reflect any existing or planned official policy or position of his employer in respect thereto.