justin0469 Posted January 15, 2010 Report Share Posted January 15, 2010 Short version:http://www.popsci.com/technology/article/2010-01/new-coding-could-make-internet-90-percent-more-energy-efficient New Coding Could Make Internet 99 Percent More Energy Efficient By Stuart Fox Posted 01.13.2010 A Series Of Tubes via The Collegian For most people, conserving energy means turning off lights in empty rooms. But for the researchers at Bell Labs, the massive energy savings lurk in the 1's and 0's of the code that regulates the Internet. Based on a new study from the lab, communications networks could use 99 percent less energy with only a few simple code changes. Bell Labs also estimates that those savings would prevent the emission of 300 million tons of carbon. To help implement the changes needed to save that energy, Bell Labs has formed Green Touch, a consortium of networking and computer companies dedicated to reducing the energy waste of telecom networks by 99.9 percent over the next five years. The savings lie in programs that separate actual messages from electromagnetic fuzz. Right now, networks use very powerful signals to communicate, that way the signal vastly overpowers the noise. Now that computer programs can easily separate the signal from the noise at much lower energies, networks can transmit the same message at a fraction of the energy cost. And this development couldn't come soon enough. With ubiquitous Wi-Fi and 3-D television right around the corner, the scale of electronic communication, and thus the scale of the energy waste, is about to increase exponentially. With easy energy-saving progress by Green Touch and similar groups, we might be able to enjoy an on-demand 3-D movie without contributing to global warming. Long version:http://www.newscientist.com/article/dn18377-worlds-communications-network-due-an-energy-diet.html?DCMP=OTC-rss&nsref=online-newshttp://www.greentouch.org/ Quote Link to comment Share on other sites More sharing options...
Disclaimer Posted January 15, 2010 Report Share Posted January 15, 2010 So basically.... instead of common +/-5V, or +/-3.3V Vpeak, they're going to lower the Vpeak closer to the noise floor and rely on software to separate the signal and noise.Lower the SNR = massive energy savings? Am I understanding this right? Quote Link to comment Share on other sites More sharing options...
justin0469 Posted January 15, 2010 Author Report Share Posted January 15, 2010 ha that's what it sounds like but it makes it sound like a dumb idea when you explain it Quote Link to comment Share on other sites More sharing options...
Disclaimer Posted January 15, 2010 Report Share Posted January 15, 2010 Everything sounds like a dumb idea when I explain it - just ask my girlfriend. Quote Link to comment Share on other sites More sharing options...
wrillo Posted January 15, 2010 Report Share Posted January 15, 2010 Everything sounds like a dumb idea when I explain it - just ask my girlfriend. I have that same problem! so true! Quote Link to comment Share on other sites More sharing options...
Fonzie Posted January 15, 2010 Report Share Posted January 15, 2010 3D TV is coming soon?!? Quote Link to comment Share on other sites More sharing options...
jarvismb Posted January 15, 2010 Report Share Posted January 15, 2010 .....or we could switch to optical transceivers and bite bullet for cost now and save big numbers with lots of zeroes in the future.....I've never understood why we didn't start whoring optical as soon as we harnessed the technology. The SNR equations barely even apply to optical like they traditionally do with conductors. The only losses are junction losses and angle refraction aggregations over long distances. For all intents and purposes, you physically can't induce noise into the system once the signal is transmitted. Hence, no more SNR problems and no more crazy auto-correlation or convolution algorithms to find your long-lost signal.... Quote Link to comment Share on other sites More sharing options...
justin0469 Posted January 15, 2010 Author Report Share Posted January 15, 2010 because ppl and companies wont "bite the bullet" and lose what theyve invested in copper/coax already Quote Link to comment Share on other sites More sharing options...
cmh_sprint Posted January 15, 2010 Report Share Posted January 15, 2010 because ppl and companies wont "bite the bullet" and lose what theyve invested in copper/coax already+1. The company I work for is spending $$$$$$ to push more bandwidth over the in-bedded cooper infrastructure rather that pushing more fiber out to the back bone. Quote Link to comment Share on other sites More sharing options...
jarvismb Posted January 15, 2010 Report Share Posted January 15, 2010 I understand why, the question was rhetorical, I was more or less venting.I won't be the first engineer to quietly shake their head while some 'business' person ignores advanced technology to save a buck then asks why it's insufficient for growth down the road and spends twice as much playing catch-up in the end. And I won't be the last.... Quote Link to comment Share on other sites More sharing options...
Disclaimer Posted January 15, 2010 Report Share Posted January 15, 2010 I agree from an engineer standpoint, but when I put on my business hat - economics doesn't favor long-term business case solutions. C'est la vie. Quote Link to comment Share on other sites More sharing options...
jarvismb Posted January 15, 2010 Report Share Posted January 15, 2010 being a person the both fiber and copper comm systems on my PCBs every day, no one will ever convince me that it was a good business idea to try and squeeze the life out of copper when we had the chance to pony up for fiber long ago. Play all the fancy number games you want with copper, you'll never get the speed out of it (for the price) that you will from optics. At least not with current tech. You can combat this POV in years to come if we have some new cool breakthrough, but as of now, it was a stupid decision made by people who didn't understand the technology they were presented. Quote Link to comment Share on other sites More sharing options...
Likwid Posted January 15, 2010 Report Share Posted January 15, 2010 Dudes, the only way you can improve the interweb efficiencies is to increase the diamedius of the tubez.Bigger tubez = less energy, simple fisics people Quote Link to comment Share on other sites More sharing options...
jarvismb Posted January 15, 2010 Report Share Posted January 15, 2010 ^^^ </thread> Quote Link to comment Share on other sites More sharing options...
yotaman88210 Posted January 15, 2010 Report Share Posted January 15, 2010 I love lamp Quote Link to comment Share on other sites More sharing options...
InyaAzz Posted January 15, 2010 Report Share Posted January 15, 2010 I love lampLinux and the holy trinity Quote Link to comment Share on other sites More sharing options...
e-flores Posted January 15, 2010 Report Share Posted January 15, 2010 My head hurts from reading this thread.... Quote Link to comment Share on other sites More sharing options...
jbot Posted January 15, 2010 Report Share Posted January 15, 2010 is chuck norris going to have to choke a geeky bitch? Quote Link to comment Share on other sites More sharing options...
InyaAzz Posted January 15, 2010 Report Share Posted January 15, 2010 Quote Link to comment Share on other sites More sharing options...
jblosser Posted January 16, 2010 Report Share Posted January 16, 2010 (edited) .... with only a few simple code changes. Bell Labs also estimates that those savings would prevent the emission of 300 million tons of carbon. To help implement the changes needed to save that energy, Bell Labs has formed Green Touch, a consortium of networking and computer companies dedicated to reducing the energy waste of telecom networks by 99.9 percent over the next five years......So, I'm not an electrical enginerd, I don't know what language Jarvis is speaking when he talks of "angle refraction aggregations" - even if I Googled I'm pretty sure I still wouldn't get it. The article says they "need to form a consortium...of companies" and it will be over the "...next five years..." that the savings would appear.But in the preceeding sentence it says that all that needs to be done are "...a few simple code changes."So if it's so simple, why not just implement the "simple code changes", realize the savings, and go on to solve some other problem? Either I'm missing something (likely), the article is ambiguous, or it's just poorly written.I think Matt has given the best answer so far...EDIT: I was right. I Googled "angle refraction aggregations" and I'm no closer to understanding it than before.... Edited January 16, 2010 by jblosser add my edit Quote Link to comment Share on other sites More sharing options...
ReconRat Posted January 16, 2010 Report Share Posted January 16, 2010 And how well does it all combine with future quantum transmissions and the current new mil-spec spinoff of ultra high compression of data stream? Just wondering...I've already seen compression ratios of about 44 to 1 in use. 44 times less data stream. And it can go a lot higher than that on the mil-spec side. Wouldn't that be a savings beyond anything else? Just wondering... Quote Link to comment Share on other sites More sharing options...
Disclaimer Posted January 16, 2010 Report Share Posted January 16, 2010 This thread will eventually be moved... my guess is it'll end up in the SAEM group. Quote Link to comment Share on other sites More sharing options...
ReconRat Posted January 16, 2010 Report Share Posted January 16, 2010 So basically.... instead of common +/-5V, or +/-3.3V Vpeak, they're going to lower the Vpeak closer to the noise floor and rely on software to separate the signal and noise.Lower the SNR = massive energy savings? Am I understanding this right?hmmm, that's what +/-3.3V was in the first place. An energy savings...Good luck lowering the threshold. But it will have to go that low or lower for quantum computing anyway. But if I understand the use of quantum, there is no bad data. Just maybe no data at all."Angular reflection aggregations" is the reflection (bouncing) of the light off the sides (angular) of the fiber during transmission through the fiber. It basically goes out of sync with the stream and becomes noise (aggregations) at the far end. There are other circumstances, but that's the basic one. Fiber has a distance limitation because of the accumulation of this noise. Amplifiers (repeaters) are used to clean and pass the data. Quote Link to comment Share on other sites More sharing options...
ReconRat Posted January 16, 2010 Report Share Posted January 16, 2010 3D TV is coming soon?!? Yes it is, but it won't be cheap. And manufacturers are reluctant. It will probably use the existing shutter specs. One of my co-workers was into it, and there's a bunch of geeks that have been experimenting with it for years. The 3D TVs won't cost much more (or shouldn't), but the 3D shutter spectacles will be pricey.Samsung is squawking about the development of 3D TV on Facebook.Samsung 3D TV on FaceBookBtw, some people can't see 3D content. I remember older 3D movies where it took a while for the 3D to kick in. You kind of had to train your brain to work with it. Newer stuff with polarizing or shutter specs are easier, but still, 4 percent to 10 percent of people can't see the 3D.Why I can't Watch 3D TV Quote Link to comment Share on other sites More sharing options...
Fonzie Posted January 16, 2010 Report Share Posted January 16, 2010 Is that our BenYen bein' choked out by ChuckyPoo?? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.