January/February 2018 issue of acmqueue

The January/February issue of acmqueue is out now


Power Management

  Download PDF version of this article PDF

ITEM not available


Originally published in Queue vol. 1, no. 7
see this item in the ACM Digital Library



Andy Woods - Cooling the Data Center
What can be done to make cooling systems in data centers more energy efficient?

David J. Brown, Charles Reams - Toward Energy-Efficient Computing
What will it take to make server-side computing more energy efficient?

Eric Saxe - Power-Efficient Software
Power-manageable hardware can help save energy, but what can software developers do to address the problem?

Alexandra Fedorova, Juan Carlos Saez, Daniel Shelepov, Manuel Prieto - Maximizing Power Efficiency with Asymmetric Multicore Systems
Asymmetric multicore systems promise to use a lot less energy than conventional symmetric processors. How can we develop software that makes the most out of this potential?


(newest first)

Matias | Tue, 02 Mar 2010 19:12:37 UTC

I was reading "A Conversation with Steve Furber" story and from there I jumped at your conversation in 2003 were you promised an update for 2008, which I did not find it yet.

I see that your ways are somehow linked, one at Apple and the other at Intel (as an Apple supplier).

I just wanted to let you know that I really enjoyed reading your conversation and that the article keep me thinking about two things.

First, near the end David says "I dont think today about loading an application onto my cellphone. But I think in the future one can do that.". Well, after 7 years and the iPhone out there, this idea certainly came to life. Now, I´m just wandering what the Apple A4 might have under its hood, Cortex A9 and all that.

Second, reading about David idea that software must start looking at power efficiency, there are some high profile programmers that are doing it. Take for example Fabrice Bellard´s new PI record were he explains "The algorithm I used (Chudnovsky series evaluated using the binary splitting algorithm) is asymptotically slower than the Arithmetic-Geometric Mean algorithm used by Daisuke Takahashi, but it makes a more efficient use of the various CPU caches, so in practice it can be faster. Moreover, some mathematical tricks were used to speed up the binary splitting." So he´s really looking at efficiency from a software perspective.

As a final though, I was a huge fan of Transmeta (the IEEE Spectrum article was great) but I think that maybe the biggest problem Transmeta had was that Linux was not that popular at that time, even with Linus in the payroll. According to the IEEE article, Crusoe was great for Linux in its native Instruction Set but needed the on the fly translation for x86 (with the extra registers and all that) and performance suffered. Nowadays 2010 looks like the year for ARM NetBooks, and there´s no need for x86 emulation, since you can recompile any Linux software out there (even OpenOffice or AbiWord) to have it at native full speed. If only all that GPL code were available for Crusoe at that time, you might get away without any x86 emulation.

I must admit I´m not very keen of VLIW, as a Sun/ORACLE employee I prefer Niagara approach to max out processor utilization. If you need single thread performance you can out some kind of scouting, but the massive thread model looks more promising to me.

rektide | Sat, 13 Feb 2010 20:51:25 UTC

"Well, great, in five years lets look back and see how we did."

Apple A4 was just released. Maybe its time to schedule in that follow up interview.

RIP PWRficient.

Leave this field empty

Post a Comment:

© 2018 ACM, Inc. All Rights Reserved.