Building an energy efficient home computer
Summary: You say you want a revolution?Energy efficient data centers are in the news again, with the EPA reporting that data centers use 1.
You say you want a revolution? Energy efficient data centers are in the news again, with the EPA reporting that data centers use 1.5% of US electricity - almost 6 million home's worth - and doubling in five years.
But what about your home data center? My house has 2 wireless routers, 3 systems, 4 monitors, 6 cores and 8 GB RAM. They don't use as much power as a refrigerator, but I'd like them to be more efficient anyway.
Measuring & comparing power use Can we cut the power requirement? We could, if we had a way to reliably benchmark power consumption across architectures. Which is what JouleSort: A Balanced Energy-Efficiency Benchmark (PDF) by Suzanne Rivoire, Mehul A. Shah, Parthasarathy Ranganathan and Christos Kozyrakis tries to do.
The benchmark of the future The authors chose a sort algorithm that would exercise the entire system:
JouleSort is an I/O-centric benchmark that measures the energy efficiency of systems at peak use. Like previous sort benchmarks, one of its goals is to gauge the end-to-end effectiveness of improvements in system components. To do so, JouleSort allows us to compare the energy efficiencies of a variety of disparate system configurations. . . . previous sort benchmarks have been technology trend bellwethers, for example, foreshadowing the transition from supercomputers to clusters. Similarly, an important purpose of JouleSort is to chart past trends and gain insight into future trends in energy efficiency.
Prototyping an energy efficient server They used the benchmark to evaluate several systems, some "unbalanced" systems such as a laptop and systems "balanced" or configured to meet the needs of the benchmark most efficiently.
They found that unbalanced CPU utilization was quite low, ranging from 1% to 26%. As a result, the system didn't accomplish much work for the power it consumed.
Since the CPU is usually the highest power component, these results suggest that building a system with more I/O to complement the available processing capacity should provide better energy efficiencies.
Ah, the irony! 40 years after the minicomputer we are back to a batch mainframe I/O-centric architecture. All things old are new again.
Design for efficiency Disks and bandwidth are critical for efficiency in this benchmark. At 15 W each, it doesn't take many SATA disks to overtake the CPU as the major power sink. The balanced system required 2 trays of 6 disks each to keep a dual-core CPU busy.
Here's the configuration of a balanced server and note the disk components.
A really efficient server
The team then built a server optimized for the benchmark. As I noted in Power pushing 2.5? drives to tipping point 2.5" notebook drives are 2-3 watts, not 10-15. And sure enough, the paper found that an energy efficient system did much better with notebook drives. The configuration:
File systems, RAM and power supplies The benchmark is a sequential sort. The authors found that file systems with higher sequential access rates more efficient. Developers, the time may not too far away when your code is measured on power efficiency as well as performance.
They also found that reducing the RAM footprint to the needed capacity raised efficiency as well. FB-DIMMs, with their 5 watt per stick penalty, are a definite efficiency target.
They also found that the winning system could use a much smaller power supply. The authors suggest that power-factor corrected power supplies are required to make energy efficient servers economic as well.
The StorageMojo take As the breadth of the paper suggests, power efficiency requires a holistic understanding of computes, I/O, software, power factors and configuration trade-offs. Some of the supercomputer folks can do this, but the average data center or home user is years away from this level of workload understanding.
Instead the research points to a few things that increase efficiency and reduce consumption across a wide range of workloads and configurations. Mobile CPUs and notebook disks are 2 likely candidates. Software effects are significant as well because widely used software affects so many systems.
Will we all be buying home servers with a half dozen disks any time soon? No. But saving power while increasing performance is always a good thing.
Comments welcome, as always. For home users power cost is not a big issue. How much more would you be willing to pay for a more efficient system?
Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.
Talkback
Can you add a price column to the first table...
These power numbers are not accurate
George, George, must you?
<br>
<blockquote>
Power Measurement: To measure the full-system AC power consumption, we used
a digital power meter interposed between the system and the wall outlet. We
sampled this power at a rate of once per second. The meter used was Brand
Electronics Model 20-1850CI which reports true power with ?1.5% accuracy. In
this paper, we always re- port the average power over several trials and the
standard deviation in the average power.
</blockquote><br>
<br>
Sounds reasonable to me.
Robin
Pay no attention to George,
blogs, very attention starved.
George is Venomous
But then again, we shouldn't be surprised. Everything he writes is with a tinge of rage. I'd really like to understand why he is that way. It's gotten to the point that when I see an angry headline on ZDNet, I know instantly it's George. And when I click on the article, sho' nuff 100% of the time, it IS George.
George, why are you such an angry man? Why is everything you write so venomous? And why do you attack people who respond to your articles?
I say let's all boycott George Ou until he tones it down and learns to be a good little boy, instead of an antagonist. You all with me?
ignore him
If he falls in the technology jungle and no one is around to hear him, does he make a sound?
It's hard to read those charts.
A better metric
benchmarks. What makes sense to me is, "watts per Karazhan hour." My home
computer is mostly used for playing WoW, so a better benchmark for me is the
power consumption of my home electronics while busy raiding Karazhan or any of
the 25 man instances.
Is WiFi more efficient than wired Ethernet? Would I be better off using a Raid 10
enclosure full of laptop drives instead of my LaCie BigDisk d2 Extreme? FireWire or
USB 2.0?
Heck, I should go hire a decent power meter and find out :)
oops...
power (for the same performance) if the extra cost of the computer was worth
more than the savings on my power bill.
Ultimately, you could add "carbon tax" onto the power bill and the computer/
peripherals I buy, and the same decision would apply - but the Carbon Tax has to
come first. Which leads to all kinds of social repercussions when you consider that
some families can't afford computers in the first place, so Carbon Tax will just be
doubling their power bill with no means to reduce their consumption below 1
fridge running all day (they're too poor for TV, and only have a line-powered
phone).
Or I could go and try to calculate the "carbon cost" of my computer and determine
what alternatives exist for me to lower the "carbon footprint" of the next one I buy
- and just work on lowering my personal "carbon footprint" as an exercise in
applied altruism.
Power Numbers?
Power factor is corrected in the better power supplies and will help your power bill. I corrected my entire home's power factor and reduced my total billable wattage by 18%. That resulted in a 24% reduction in power cost because of our usage surcharge.
RE: Building an energy efficient home computer
HOME computer???
Of course, the ridiculous aspect of the suggestion is that within 6 months the study would be obsolete when someone came out with SOME new item that saved a few watts.
HELLO! It's a [i]HOME[/i] computer!
Years ago (late '70s) I planned on building a [b][i]really[/i][/b] hot audio system. (This was back when you still could build amps with better performance than most commercial audiophile stuff.) I bought transformers, capacitors, etc., for a 1KW supply, amp modules, etc.
What finally shocked me into seeing reality was when I seriously started trying to figure out how to build a SPECTRUM ANALYZER so I could have a really accurate idea of distortion levels.
I realize these blogs are aimed at techno/geek types, but some of these articles really go off the deep end! Like using RAID in home systems!
(Okay, if there are roaches near the computer, that's another story ....)
Nature of evolution
I think the author, in this case, is simply trying to wake up those who know how to tinker with this in their garages to start doing this. In truth, it's beyond me. As more of these super-uber geeks come up with solutions and start posting it, commercial outfits will take notice (if it grows in popularity) and incorporate it into their own products.
When you think about it, if 1.5% of U.S. power consumption comes from the data center, what percentage comes from home users' computers? There is a rather slow movement going on right now to green our devices. But as it starts to pick up steam, and consumers themselves start to demand greener products, these commercial outfits will turn to the solutions posted online. The very solutions posted by people that this author is trying to wake up in the first place.
So I commend the author for his efforts, though I totally agree with you that much of what is being discussed here and in other articles is going over the heads of the average reader.
But I am excited to know that there is someone out there who WILL read this, be inspired to come up with solutions, and those solutions will trickle down to my home devices (not just computers, but everything with a chipset in it) and help me cut back on my electric bills! I am tired of relying on that windmill in my backyard to generate electricity. It's so big, I have no room for bbq parties anymore!
Red-neck RAID Solution
That way I don't often run into automatic copying of malware to the backup, and if one drive fails, I can do a quick BIOS switch to boot from the backup while I wait for the replacement to arrive.
roach motel
Then there's all that overclocking stress...
Pretty good, but processor is too pricey.
There is delay between testing
focus on .<br>
<br>
Robin
I figured that.
AMD TL-37!
I went this route instead of a $450 laptop because laptop drives and fan is less reliable, and all the components cost more. A low power display comes with, though.
A little light on spec.
I would have went with a E4300 Core Duo and some Asus mobo with not too many features except with ICH8R controller. I also would go with 2 500gig Seagate Barracudas commercial grade and DVD for backup. A little more expensive but more durable for the data.