-
-
Categories
-
Data Center Design:
Construction,
Container,
Data Center Outages,
Monitoring,
Power and Cooling
Policy: Cap and Trade, Carbon Footprint, Carbon Reduction Commitment, Carbon Tax, Emissions
Power: Biomass, Fossil Fuel, Fuel Cell, Geothermal, Hydro, Nuclear, Solar, Wind
Application: Cloud Computing, Grid Computing
Technology: Microblogging, Networking, Servers, Storage, Supercomputer
-
Delayed gratification on new energy-saving chips - by Doug Mohney
Views and Opinions on Green IT (Sep 14 2011)
-
Intel’s Ivy Bridge chips may be delayed for up to three months, according to a couple of reports kicking around the web at the beginning of the month. And I want to pull my hair when I see the latest hype news on “revolutionary breakthrough” technologies that have been demoed in the lab – but take years to decades to see the commercial light of day.
According to the rumor mill, Intel was originally scheduled to ship Ivy Bridge by the end of this year. Instead, the company won’t be shipping until March 2011, due in part to the slumping demand for PCs around the world. Moving to a 22 nanometer (nm) process, the latest and greatest in chip tech, is expensive, so there is allegedly no rush to move up.
Ivy Bridge would save energy in two ways. First, because it’s a smaller “process,” you get either better performance at the same chip real estate or lower power with the same performance ratings. Secondly, Ivy Bridge will incorporate the company’s 3D Tri-Gate transistors to further increase performance/lower power.
A three month delay in the new chip set sounds like a big deal. For companies planning new product launches around Ivy Bridge, such as Apple, it isn’t exactly the news they were hoping for. As to the rest of us, so long as the delay doesn’t push the purchase of more energy-efficient servers into the next fiscal year, we can live with it.
The latest hype of new chip technology comes out of the University of California at Berkeley. Researchers have built an experimental ferroelectric-based material to lower the amount of voltage required to store a charge. Right now, you need about 1 volt to operate a chip and this requirement has been steady for about a decade. With the new tech, you can use less voltage and generate less heat, so you’d get a ripple effect by lowering both energy consumption for servers and cooling.
BUT—
Just because a discovery gets published and webizied doesn’t mean it will show up on chip production lines next year. Intel first disclosed its Tri-Gate technology in 2002 and is only now getting around to making chips with it. Add another 5 to 10 years of various research efforts before that, and it’s easy to get quickly depressed on the latest Big Announcements in chip technology.
Hordes of new announcements are made every year, but very few see the light of day. Memristors were first discussed in a 1971 research paper. HP actually produced one in 2008 and the low-voltage memory chips could replace hard drives, saving energy and operating at higher speeds. HP may make them commercially available by 2013. Maybe.
Login to comment.
Related Articles
- Iceland: Calm, Cool, Collected by Tate Cantrell
- also published in Views and Opinions on Green IT
- Virtues of Virtual by Carol Wilson
- also published in Views and Opinions on Green IT
- Grading Google's carbon neutral claims
- also mentions Apple
- Beyond Going Virtual by carol wilson
- also published in Views and Opinions on Green IT
-







Recent Comments
ControlCircle » Gartner: Build your own datacentre rather than hosting
It’s startling that in today’s volatile environment Gartner is prescribing such a high risk strategy. ...
Carbon3IT Ltd » Does efficiency matter when your power is renewable (and affordable)? - By Peter Judge
Peter, do you really think that this is good practice?, as you say its like ...
See all recent comments