Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

10 IT Power-Saving Myths Debunked

timothy posted more than 5 years ago | from the replace-all-leds-with-cfls dept.

Power 359

snydeq writes "InfoWorld examines 10 power-saving assumptions IT has been operating under in its quest to rein in energy costs vs. the permanent energy crisis. Under scrutiny, most such assumptions wither. From true CPU efficiency, to the life span effect of power-down frequency on servers, to SSD power consumption, to switching to DC in the datacenter, get the facts before setting your IT energy strategy."

cancel ×

359 comments

Sorry! There are no comments related to the filter you selected.

I dunno.. (5, Interesting)

Anrego (830717) | more than 5 years ago | (#25286499)

I'm of the school that thinks "debunking" involves some kind of comprehensive stats or numbers or evidence weight against strongly held opinions.

This article is basically a verbose version of the "nuh uh" argument.

It's not a bad article.. but I would hardly call this "debunking".

And I totally disagree on point #2 .. maybe having _all_ your extra servers always on is bad.. but if load peaks there is no _way_ someone should be waiting while a system boots.

Re:I dunno.. (4, Informative)

Nursie (632944) | more than 5 years ago | (#25286583)

That depends if your system has been tuned to boot in 5 seconds.

Or if it can return from suspend-to-ram nice and quick.

Single page (4, Informative)

gnick (1211984) | more than 5 years ago | (#25286733)

Sorry for the thread hijack, but I decided to post this link as soon as I saw the links to all 4 pages of the top 10 list.
http://www.infoworld.com/archives/emailPrint.jsp?R=printThis&A=/article/08/10/06/40TC-power-myths_1.html [infoworld.com]

Re:Single page (1, Insightful)

Anonymous Coward | more than 5 years ago | (#25287009)

Nonetheless, it's off topic to what we're discussing in this thread. Pretty much anyone can click the 'print' button that's on the article itself. Way to go with the karma whoring though.

Re:I dunno.. (1)

thePowerOfGrayskull (905905) | more than 5 years ago | (#25286919)

That depends if your system has been tuned to boot in 5 seconds.

Or if it can return from suspend-to-ram nice and quick.

Actually it doesn't. When you're in an HA environment, that means /being/ HA - there must be /no/ wait time while a machine starts (even if it's only a couple of seconds).

Re:I dunno.. (3, Interesting)

Nursie (632944) | more than 5 years ago | (#25286985)

Think past "HA" for a second.

Think about metrics, predictable traffic and planned capacity.

Think about bringing a percentage of spare capacity online at any one time, in line with predicted peak traffic, and more as the load increases on what's there already.

HA can still be HA without needing everything on all the time.

(also, why the hell was my last post modded down as redundant?)

Re:I dunno.. (2, Insightful)

cparker15 (779546) | more than 5 years ago | (#25287277)

(also, why the hell was my last post modded down as redundant?)

Probably because a similar point was already made in TFA:

You can also select systems that cold-boot rapidly. Model to model and brand to brand, servers exhibit wide variances in power-up delay. This metric isn't usually measured, but it becomes relevant when you control power consumption by switching off system power. It needn't take long. Servers or blades that boot from a snapshot, a copy of RAM loaded from disk or a SAN can go from power-down mode to work-ready in less than a minute. The most efficient members of a reserve/disaster farm can quiesce in a suspend-to-RAM state rather than be powered down fully so that wake-up does not require BIOS self-test or device querying and cataloging, two major sources of boot delay.

Re:I dunno.. (1)

zippthorne (748122) | more than 5 years ago | (#25287369)

Which thereby "debunks" point one: the power-on statistics obtained in a cold boot will not be present if they are not run in a hibernate-power-up. But the effects will still be there, if any, because the temperature still cycles.

Re:I dunno.. (4, Informative)

Anonymous Coward | more than 5 years ago | (#25286641)

If you're booting those servers diskless with PXE and NFS, the boot time should be negligible. I should imagine the trick would also be to bring additional resources online before you are the point that you must tell users to wait while the server boots. The magic would be in predicting near-term future use...

You mean ... (4, Funny)

khasim (1285) | more than 5 years ago | (#25286959)

... something like monitoring system usage and bringing additional boxes up when usage hits something like 80%?

And then suspending boxes when usage drops down to 10%?

All in all, trying to maintain a level 50% utilization level? Maybe with the utilization level setting being an option that the sysadmin could change?

I'd recommend you patent that idea.

Re:You mean ... (1)

Nursie (632944) | more than 5 years ago | (#25287073)

I'd reckon IBM and VMWare probably have that lot wrapped up already. Still, there's no reason (given current record) that you couldn't have one as well.

Re:You mean ... (2, Informative)

Sobrique (543255) | more than 5 years ago | (#25287787)

Actually you're pretty close with VMware at the moment - VM instances can be 'hot' migrated, so you can clump them up on one server, power the rest down, and fire them up/migrate when demand shows. Your response won't be great, but at least you will be able to respond to dynamic load fluctuation.

Actually VM tech goes a long way to doing that anyway, provided you've a vaguely good concept of workload fluctuations.

Sod NFS (3, Interesting)

Colin Smith (2679) | more than 5 years ago | (#25287713)

Sorry, It's just not worth the pain. Boot to RAM.

You just set high and low load thresholds for server on/off. And a load balancer which simply adds the new server to the server pool when it notices it's there, removes them when it's gone. So no need to try to predict stuff.

5 seconds or 3 minutes, the server boot times are largely irrelevant. If you think you're going to handle a slashdotting you are mistaken, you can't handle oneoff events this way. You would have to go from 1 to 100 servers and connections in 5 seconds.

What it can do is grow really quickly if a service becomes very popular very quickly, or reduce your datacenter costs if it's typically used only 9-5. Or even, dual purpose processing. Servers do X from 9-17 and Y from 15-20.

 

Re:I dunno.. (5, Informative)

gnick (1211984) | more than 5 years ago | (#25286863)

FTA:

Hibernate mode in XP saves the state of the system to RAM and then maintains the RAM image even though the rest of the system is powered down.

They must be using a different version of XP than I am... When I 'Hibernate' my laptop, it dumps the RAM to a file on the hard drive and then powers off completely. When I 'Stand By' my system, it keeps everything in RAM.

Maybe they have SP4...

Re:I dunno.. (5, Funny)

halcyon1234 (834388) | more than 5 years ago | (#25287245)

hey must be using a different version of XP than I am... When I 'Hibernate' my laptop, it dumps the RAM to a file on the hard drive and then powers off completely.

You must be using a different version of XP than I am... When I 'Hibernate' my laptop, it attempts to dump the RAM to a file, throws a hissy fit like a coddled freshman after their first exam, fails miserably, flickers the screen, disables the Hibernate option, and then just sits around until the battery drains.

Re:I dunno.. (5, Interesting)

ShieldW0lf (601553) | more than 5 years ago | (#25286865)

I've got electric heat, and I've got a pile of servers in my spare bedroom, and I never need to turn on the electric heat, because the servers heat my home.

Which looks to me like an opportunity. People pay for heat. So, put the servers where people need heat, and suddenly a liability is a resource.

Apartment buildings, office buildings and malls in cold climates should all be prime locations for a datacenter.

Re:I dunno.. (5, Informative)

snowraver1 (1052510) | more than 5 years ago | (#25287089)

If you are using electric heat, chances are you don't live in a cold climate and pay for air conditioning for much of the year negating any "savings". Here in cold-balls Canada, EVERYONE has centeral heating; it's too expensive to use electricity. That being said, I do agree that datacenters' heat should be used to heat useful things (office bldgs, like you suggest).

Re:I dunno.. (2, Informative)

Aliencow (653119) | more than 5 years ago | (#25287239)

Well I live in Canada, and most people I know use electric heating. Yes, central electric heating is great, and actually cheaper than oil around here. (Montreal area)

Re:I dunno.. (4, Funny)

geminidomino (614729) | more than 5 years ago | (#25287643)

Hell, the way things are going, soon hiring a cadre of hookers to rub on you for heat will be less expensive than oil.

Re:I dunno.. (1)

Vagnaard (1366015) | more than 5 years ago | (#25287279)

I don't know from where in Canada you are but no one I know in Montreal has central heating. Even in big apartment buildings. Everybody use that expensive electric heating. Maybe that's because electricity is cheap in Quebec...

Re:I dunno.. (1)

snowraver1 (1052510) | more than 5 years ago | (#25287679)

From Calgary. I pay 7 cents/ kWh for electricity. I don't know a single person that DOES use electricity for heating. It's all natural gas here.

What is the electricity rate in Montreal/Quebec?

Re:I dunno.. (2, Interesting)

greenzrx (931038) | more than 5 years ago | (#25287365)

back in the very early 90's We moved into the first 7 world trade center. There was no heat in that building, it was air conditioned 365 days a year. People and computers heated all the floors. Unoccupied floors were damn cold in winter, i can tell you.

Re:I dunno.. (1)

camperdave (969942) | more than 5 years ago | (#25287355)

I have no control over the heating in my apartment. Thus, I have to have an air conditioner running 24/7/365 because my room gets too hot from the two computers I have running... and they're both sub gigahertz machines. Heaven help me if I upgrade to a real machine.

Loving Myth #2 tip... (1)

PinkyDead (862370) | more than 5 years ago | (#25286939)

If you have a shit system that's really slow and badly written display the following:

"The sub-optimal response you are experiencing will soon be resolved as we are utilising quantum replicators to produce more server hardware for your request. Once complete we will travel back in time and resubmit your request. Thank you for using One-Born-Every-Minute hosting. Have a nice day."

(Speaking from experience - different text, same message).

Re:I dunno.. (5, Insightful)

R2.0 (532027) | more than 5 years ago | (#25287197)

I stopped reading at #1: "Fact: The same electrical components that are used in IT equipment are used in complex devices that are routinely subjected to power cycles and temperature extremes, such as factory-floor automation, medical devices, and your car."

Well, yes, except for the fact that the it's a total lie. Cars, factory automation, and medical devises most certainly do NOT use "the same" components. While they may do the same things, and even be functionally equivalent, they are rated to much higher temperature and stress levels than consumer or even server grade components. Just ask the folks who have been trying to install "in-car" PC's with consumer grade components.

Re:I dunno.. (-1, Troll)

Toll_Free (1295136) | more than 5 years ago | (#25287385)

You're right.

They don't use Transistors, Inductors and Capacitors at all, do they?

Functionally, the devices might not be used to perform the same precursory things... However, silicon is silicon, capacitors are still made from the same things, and inductors are little more than (get this!!!) lumps of a constant controlled by the turns ratio and other things.

Nope, totally different animals, huh?

--Toll_Free

Re:I dunno.. (4, Insightful)

PlusFiveTroll (754249) | more than 5 years ago | (#25287797)

However, silicon is silicon, capacitors are still made from the same things

Thank you for playing the game, but you have lost [badcaps.net] . Rather then using more expensive Nippon electronics, the Chinese parts you used had a few part per million more impurities. This lead to early thermal failure of your mainboard.

If you would like to play the game again, please acquire more venture capital and buy quality next time. You may still lose the game to your manufacture buying counterfeit parts, using the wrong specification solder, or unforeseen interactions from running at many gigahertz at high temperature.

This show has been hosted by an automation robot that costs 75 times what your laptop does and still has occasional electronics failures. :)

Re:I dunno.. (0)

Anonymous Coward | more than 5 years ago | (#25287367)

Ok, I really tried to RTFA but only made it as far as Myth 2, where he says you don't need to leave backup servers live, and when the customers complain tell them you are 'bringing more resources online'.

Like a million WOW players give a half a shit WHY it suddenly takes 5 minutes to transition into the next play zone, they just are going to be pissed your servers react slow.

If you REALLY want to know what works & what doesn't, get an outlet meter, and just try different setups and track the actual usage.

Re:I dunno.. (4, Funny)

sorak (246725) | more than 5 years ago | (#25287403)

For a Web site, put up a static page asking users to wait while additional resources are brought online.

We're sorry for the inconvenience, but our systems seem to have been shut down. We've asked leroy, rufus, and heraldo to hit the power button, and we assure you that, once they've found that button, they will push it, and then, once the mandatory scandisk operation has completed, the Windows server screen will appear, and once the kernel operations have completed, the services you have requested will be available.

And that will be awesome!

While you're waiting, here are some links to our competitors' sites. Remember to open them in a new tab, so you can occasionally come back and hit "refresh". We promise, we're almost ready to serve you.

Re:I dunno.. (3, Informative)

SuperQ (431) | more than 5 years ago | (#25287723)

Yea, I don't know who wrote that bit in the article, but they're just dumb. If you run any kind of system with a load balancer in front of it you can easily script starting up additional machines as soon as your monitoring says you reach 90% capacity.

Is it true... (-1, Troll)

Adolf Hitroll (562418) | more than 5 years ago | (#25286503)

...that Cowboy neal shits on his desktop to kill the cockroaches?

Sleep != Hibernate (5, Informative)

Taimat (944976) | more than 5 years ago | (#25286575)

Myth No. 6: A notebook doesn't use any power when it's suspended or sleeping. USB devices charge from the notebook's AC adapter. Fact: Sleep (in Vista) or Hibernate mode in XP saves the state of the system to RAM and then maintains the RAM image even though the rest of the system is powered down. Suspend saves the state of the system to hard disk, which reduces the boot time greatly and allows the system to be shut down. Sleeping continues to draw a small amount of power, between 1 and 3 watts, even though the system appears to be inactive. By comparison, Suspend draws less than 1 watt. Even over the course of a year, this difference is probably negligible.

um... Hibernate != Sleep. Hibernate in XP saves the RAM to the Hard Drive, and powers off. Suspend keeps RAM powered....

Re:Sleep != Hibernate (1)

gEvil (beta) (945888) | more than 5 years ago | (#25286709)

Mod up! The article got this one completely backwards.

Re:Sleep != Hibernate (5, Interesting)

EvilRyry (1025309) | more than 5 years ago | (#25286905)

Using my handy killawatt, I tested how much power my desktop (not including accessories) draws while off, on and idle, on and under load, and in S3 suspend.

Off - 6 watts
Idle - 140W (dropped from 152W after installing a tickless kernel)
Loaded - 220W
S3 - 8 watts

Ever since I ran that test, I put my machine into suspend at every opportunity. 140W is a lot of juice in the land of $0.18/kWh.

Re:Sleep != Hibernate (5, Informative)

afidel (530433) | more than 5 years ago | (#25286977)

How about build an energy efficient PC! I have a LP AMD 64 x2 with a Geforce 7600GS, 2 HDD's, 2GB of ram and a TV tuner and an 85% efficient PSU and I peak at around 150W, using 140W at idle is insane. For the next generation of games I'm thinking about upgrading to a 9600 GSO but that will up my idle and peak numbers by at 20W so I'm holding off till I get a game that really needs it.

Re:Sleep != Hibernate (4, Interesting)

Volante3192 (953645) | more than 5 years ago | (#25287337)

My kingdom for a mod point...

I built a new system in July, Intel Core2Duo E8400, 2gb ram, ATI 3850, two hds (one's a raptor), and the box on idle pulls 81W.

My old box, an Athlon 1800+ (actual speed: 1350hz), 2gb ram, two HDs...idle was in the 130s 140s.

(Both are excluding monitor, a 20" LCD with pulls 35-40W)

So not only did I build me a faster system, it's nearly half as power hungry as my old box.

Re:Sleep != Hibernate (0)

Anonymous Coward | more than 5 years ago | (#25287849)

heh that's funny...
I have a AMD64 X2 with 2GB RAM and a 8800 GTS, 3 HDDs, some PCI cards (sound card, wlan) etc.
Now guess what the multimeter says?
120W in idle. The funny thing is, my PSU is about 4 years old, mind you it wasn't "cheap",
but I don't think it goes anywhere near 85%+.

Re:Sleep != Hibernate (1)

lorenzo.boccaccia (1263310) | more than 5 years ago | (#25287221)

a pc office does 8 hour at day for 200 days at years. so:

0.14kW *0.18$/kWh * 8*200h = 40$? at years? of electricity? not a whole saving you could get from reducing that. or I'm doing it all wrong?

Re:Sleep != Hibernate (1)

Toll_Free (1295136) | more than 5 years ago | (#25287453)

The problem is your only looking at baseline power.

Your power rates go up as you use more power daily, weekly and monthly. Some utilities can also charge you depending on the time your using the power.

So, your answer is VERY simplistic, but shows a BASE cost of running that computer, if you have NOTHING else on during that time.

Actual expenses, in my experience, are two to three times what your posing, just due to the fact that electricity rates can double or triple post base-line power.

--Toll_Free

Re:Sleep != Hibernate (1)

lorenzo.boccaccia (1263310) | more than 5 years ago | (#25287669)

nevertheless, even 200$ at year are not a major expenses, and a 10% power saving is not so cost/effective.

I'm reversing the reasoning of demand/cost relationship: if the cost is so low, what is this energy crisis we are hearing about (I know the facts, but I fully expected a higher cost for energy given the situation)?

Re:Sleep != Hibernate (1)

Kjella (173770) | more than 5 years ago | (#25287685)

0.14kW *0.18$/kWh * 8*200h = 40$? at years? of electricity? not a whole saving you could get from reducing that. or I'm doing it all wrong?

Something's wrong about the units but I think the figure is right. Of course, it also depends on what climate you're in, if you're running an AC to get rid of those 140W again that costs too, while here up north a lot of the time it supplements other heating in the winter so it actually costs less. Anyway I haven't bothered to check but I imagine my 42" TV draws more when it's on so it's not like the computer is the big sinner. For me lower power is about less fans and less noise, there's no way 140W draw is silent.

Re:Sleep != Hibernate (1)

cpghost (719344) | more than 5 years ago | (#25287687)

140W is a lot of juice in the land of $0.18/kWh.

It's even a lot more in Europe, at approx. EUR 0.25/kWh ($0.34/kWh)...

Re:Sleep != Hibernate (0)

Anonymous Coward | more than 5 years ago | (#25286949)

That's right.

Sleep is something I do at the keyboard....*snore snore snore*

Kitteh pr0n (0, Troll)

Hognoxious (631665) | more than 5 years ago | (#25286627)

As for the wait, people will stay on hold if they know their call will be answered.

You need to keep them amused while they're waiting. Show them some nice pictures of kittens. Or some pr0n.

Re:Kitteh pr0n (4, Funny)

rugatero (1292060) | more than 5 years ago | (#25286773)

Show them some nice pictures of kittens. Or some pr0n.

I, for one, was very relieved to see the word or.

Re:Kitteh pr0n (1)

Hognoxious (631665) | more than 5 years ago | (#25287607)

Not half as relieved as the cats were.

For mere mortals there is speed (2, Interesting)

houghi (78078) | more than 5 years ago | (#25286665)

I lock my PC at the evening and turn off my monitor. Shutting down takes 5 minutes. Starting up takes 15 minutes. Just checked those time this morning to talk about it to IT. This does not include logging into the remote system with Citrix that takes another 10 minutes.

So the company has a choice.
1) Pay me (and everybody else in the company) 20 minutes
2) Pay the electricity for not turning of the PC
3) Find a solution that makes it possible to do all of this faster.

Oh. The only reason we use Citrix is to run Outlook in it.

Re:For mere mortals there is speed (1)

icebrain (944107) | more than 5 years ago | (#25286801)

Fifteen minutes to start up?! What the heck are you running on there?

We can cold-boot into XP with novell, start Notes, Catia, Enovia, and everything else in less than five. Ten after a power outage and the server connection needs to be reestablished.

Re:For mere mortals there is speed (2, Insightful)

Number6.2 (71553) | more than 5 years ago | (#25287017)

What are they running? Corporate Crippleware: Safe Boot. Virus Checkers. Keyboard Loggers (Hi, guys!). After a few "Regime Changes" it all adds up...

Re:For mere mortals there is speed (2, Insightful)

modmans2ndcoming (929661) | more than 5 years ago | (#25287069)

lots of companies have lazy desktop admins who write one giant script to run that checks every resource available in the system for every user even though most of those users will not use most of the resources it is checking for. Smart companies have created multiple scripts and figured out smart ways to quickly identify what scripts the logging in user needs to run, thus reducing boot up significantly.

Re:For mere mortals there is speed (1)

youngerpants (255314) | more than 5 years ago | (#25286867)

Option 3 looks good to me.

"Oh. The only reason we use Citrix is to run Outlook in it."

Outlook Web Access (OWA). There, I've just saved you 10 minutes per day, or around 60 hours per year. I don't know how many members of staff your company has, but if you're using Citrix, I'd imagine quite a lot... lets say 500.

500 x 60 = 30000 man hours per year saved by switching on OWA and NATing port 443.

Re:For mere mortals there is speed (0)

Anonymous Coward | more than 5 years ago | (#25287047)

"Oh. The only reason we use Citrix is to run Outlook in it."

Outlook Web Access (OWA). There, I've just saved you 10 minutes per day

OWA has a very different feature set than Outlook, especially if you're not using internet explorer. Your solution won't work in all cases.

Re:For mere mortals there is speed (1)

swb (14022) | more than 5 years ago | (#25287431)

But OWA has a much thinner feature set, and over time, they will spend more time using OWA than waiting for Citrix.

Not only that. (2, Interesting)

khasim (1285) | more than 5 years ago | (#25286871)

We have a similar problem here that I've not been allowed to fix yet.

The employees typically turn on their computers and then LEAVE THE OFFICE to get Starbucks coffee or whatever. A 10 minute wait turns into 30 minutes of non-productivity.

The computers should be the same as the phones. Instant on - any time - every time.

Re:Not only that. (1)

modmans2ndcoming (929661) | more than 5 years ago | (#25287115)

get computers that suspend to flash. Then you can turn it on and it instantly is working with a nice little unlock screen.

Re:For mere mortals there is speed (1)

afidel (530433) | more than 5 years ago | (#25287031)

Um, your computer is way underpowered or your IT department sucks because 15 minutes to boot in crazy. I have an old T42 with a 4200 rpm HDD and it only takes about 5 minutes to boot and that's with multiple server type services installed (I have two copies of MSDE installed if that tells you anything). Also Citrix logon times at my shop are ~90 seconds average and they will be more like 30 once I get the users profiles onto a faster file server.

Re:For mere mortals there is speed (3, Funny)

RMH101 (636144) | more than 5 years ago | (#25287217)

It's possibly a combination of the two. My old work laptop (Tosh Centrino, 1.6 or 1.8GHz, 1GB RAM, Win2K) used to take around 12 minutes to boot from cold. Quite a bit of this is due to the Pointsec full disk encryption software, followed by SAV, followed by the usual corporate crippleware. Horrible. In the end it became a tethered desktop as I couldn't be bothered taking it anywhere.

Re:For mere mortals there is speed (0)

Anonymous Coward | more than 5 years ago | (#25287641)

Certain places have a lot of security scripts that run when the machine boots, along with network connections and such. My machine takes between 5-15 minutes to boot on a regular basis. If it's been physically disconnected from the network, it's usually 15 minutes.

Re:For mere mortals there is speed (0)

Anonymous Coward | more than 5 years ago | (#25287281)

Exchange supports pop/imap so every mail program out there but mail.app would work fine.

OWA is also good.

Finally, 15 min boot time is just plain FAIL unless its a server. The company needs to refresh hardware or at least add RAM so machines are not forced into swapping. At the minimum, if no Windows specific stuff is needed, one can go redhat or suse for the corporate desktop.

Re:For mere mortals there is speed (2, Funny)

Bobfrankly1 (1043848) | more than 5 years ago | (#25287597)

You must be the guy who bought my old 386 66. You need to hit the turbo button to get it to boot faster!

Single Page Link (0, Redundant)

Anonymous Coward | more than 5 years ago | (#25286693)

Link to 1 page version. [infoworld.com] Anonymous so as not to be a karma whore.

Re:Single Page Link (0)

Anonymous Coward | more than 5 years ago | (#25286829)

-1 redundant, the karma whore got there before you :-)

Questionable grasp on the problem space. (5, Insightful)

Anonymous Coward | more than 5 years ago | (#25286703)

Myth No. 3: The power rating (in watts) of a CPU is a simple measurement of the system's efficiency.
Fact: Efficiency is measured in percentage of power converted, which can range from 50 to 90 percent or more. The AC power not converted to DC is lost as heat...Unfortunately, it's often difficult to tell the efficiency of a power supply, and many manufacturers don't publish the number.

I'm not sold on taking advice who doesn't understand the difference between the wattage rating of a CPU and the wattage rating of the power supply. They're completely different components.

Re:Questionable grasp on the problem space. (2, Funny)

Vagnaard (1366015) | more than 5 years ago | (#25286815)

CPU , PSU are near. This might have been an abbreviation error.

Still , shame on them for not editing their stuff properly (If they meant PSU)

Re:Questionable grasp on the problem space. (5, Informative)

Anonymous Coward | more than 5 years ago | (#25286845)

I like how this plays with the following assertion filed under "Myth No. 9: Going to DC power will inevitably save energy."

"New servers have 95 percent efficient power supplies, so any power savings you might have gotten by going DC is lost in the transmission process."

So, when it suits his argument, power supply efficiencies range from 50-90% efficency, and are kept hidden by manufacturers. Then, when that doesn't suit his argument, all of a sudden power supplies are at least 95% efficient, and everyone knows that.

I call shenanigans!

Re:Questionable grasp on the problem space. (1)

Guysmiley777 (880063) | more than 5 years ago | (#25287519)

It's called 'moving the goalposts'. It's like the author decided he was going to write about how wrong everyone was about power usage in computers, then found out that the myths were actually correct.

Oops.

Re:Questionable grasp on the problem space. (5, Insightful)

gmack (197796) | more than 5 years ago | (#25286923)

Myth No. 9: Going to DC power will inevitably save energy.
Fact: Going to DC power entails removing the power supplies from a rack of servers or all the servers in a datacenter and consolidating the AC-DC power supply into a single unit for all the systems. Doing this may not actually be more efficient since you lose a lot of power over the even relatively small distances between the consolidated unit and the machines. New servers have 95 percent efficient power supplies, so any power savings you might have gotten by going DC is lost in the transmission process. Your savings will really depend on the relative efficiency of the power supplies in the servers you're buying as well as the one in the consolidated unit.

This is completely wrong. The author missed out on two of the three power conversions that take place in a data center. Data center UPS units take the AC current convert to DC then back again just so the server can convert it back to DC. Even if you have 95% efficiency at each stage the conversion losses will add up.

People wouldn't be going DC if it didn't result in measurable power savings.

Re:Questionable grasp on the problem space. (0)

Anonymous Coward | more than 5 years ago | (#25287359)

The server is still converting to DC even if you use a DC power system; you don't really think a server can use -48V directly, do you?

Re:Questionable grasp on the problem space. (3, Insightful)

SuperQ (431) | more than 5 years ago | (#25287847)

But all machines do that anyway. Ram runs at 1.5V or 1.8V, the CPU runs at 1.2ish these days. Where does that come from? 3.3V or 5V rails..

This is why people are moving everything to the 12V rail on the PSU (ATX12V standard and other ideas) A single efficient conversion with a local on-board conversion is best.

DC power still has a lot of other issues.

Re:Questionable grasp on the problem space. (1)

aaarrrgggh (9205) | more than 5 years ago | (#25287841)

With the US power system, you do avoid four high loss (DC:AC or AC:AC) power conversions and replace them with two lower loss (DC:DC) conversions. But, compared to the ROW electrical systems, you only save two AC:AC conversions which will just gain you two points or so.

I like 600VDC as a solution, but it will only work well for the biggest consumers where you can justify a significant increase capital cost with the energy savings. It's nice to have a single 4.8MW critical power bus (with a couple spares)-- makes load management much easier and reduces space requirements significantly.

Re:Questionable grasp on the problem space. (1)

Penguinoflight (517245) | more than 5 years ago | (#25287379)

Didn't they learn you that the CPU is the big box with the blinkin' lights?

They got hibernate and suspend backwards!!! (0)

Anonymous Coward | more than 5 years ago | (#25286723)

Don't know about this but when someone gets confused about simply definitions, I start to wonder. I don't know if I can really trust what is "myth" or "fact". Or did MS redefine what hibernate and suspend, because that is what it is "in Vista".
"Sleep (in Vista) or Hibernate mode in XP saves the state of the system to RAM and then maintains the RAM image even though the rest of the system is powered down. Suspend saves the state of the system to hard disk"
Wiki http://en.wikipedia.org/wiki/Hibernate_(OS_feature)
"Hibernate is a feature seen in many operating systems where the contents of RAM is written to non-volatile storage, such as the hard disk "

Debunk this (5, Insightful)

sargeUSMC (905860) | more than 5 years ago | (#25286737)

Taking ten suppositions and making suppositions about those suppositions (I'm getting dizzy) is not debunking. All I see here is lots of questionable, completely unattributed information. For example: "The average 17-inch LCD monitor consumes 35 watts of electricity". Really? Where did this information come from? Did you pull this information from the glossy for a 17" monitor? Did you just test your monitor? Did you test a large sample of monitor's here? Did you pull this information from a study? Out of your ass?

Re:Debunk this (1)

Penguinoflight (517245) | more than 5 years ago | (#25287141)

At the risk of sounding like an idiot, that is a fairly accurate guess for 17" lcds (TN panels anyway).
Since pretty much all 17" lcds are TN rather than high-contrast panels, it doesn't really hurt to generalize.

Power ratings on monitors aren't like the ratings on computer power supplies. By effectively estimating average and peak power draw the manufacturer can save money. If the AC adapter is rated to handle too little power, the adapter or monitor will prematurely fail. If the adapter is rated too highly, the vendor just wasted a few cents.

Re:Debunk this (4, Funny)

halcyon1234 (834388) | more than 5 years ago | (#25287325)

They have a 17" monitor up their asses? Well, good to know goatse guy is getting steady work these days, even if he isn't well-versed in the scientific method.

Power (4, Funny)

Windows_NT (1353809) | more than 5 years ago | (#25286881)

Turning off your computer is always a good time to give the hamsters food and water, lets them rest, so in the morning your computer will be nice and fast. If it takes parents computer 15 minutes, his hamster need less weight

Did I miss something? (4, Insightful)

Taibhsear (1286214) | more than 5 years ago | (#25286911)

Did the definitions of 'fact' and 'debunk' change recently? Every 'myth' listed has 'fact' under it proving it is true. According to my good friend Mr. Webster this is called 'confirmation.'

Re:Did I miss something? (0)

Anonymous Coward | more than 5 years ago | (#25287149)

You misread, the "Facts" are counterpoints to the myths, proving them wrong.

Re:Did I miss something? (4, Funny)

camperdave (969942) | more than 5 years ago | (#25287631)

Yes, they've been redefined by Messrs Hyneman and Savage. Basically, if something takes a ridiculous amount of effort to blow up, then it is debunked, or "busted". If it blows up without too much provocation, it is "confirmed". If it merely catches fire, it is "plausible".

Some things conveniently left out (1, Informative)

storkus (179708) | more than 5 years ago | (#25286963)

1. Spinning up and down hard drives: as discussed in plenty of places, including here on /. I believe, you can dramatically reduce the life of drives when you cycle them due to mechanical wear-and-tear. Laptop drives are designed for a lot more cycles because they're intended for this, but if you do it constantly even they'll die sooner or later. Server and desktop drives, OTOH, will die MUCH sooner. Is it really worth some extra power in a server farm in exchange for dying drives and their associated cost, including the increased possibility of data loss?

2. LCD Backlighting: Same as above--cycling power on any kind of discharge lamp dramatically reduces its life. And while LED backlighting is VERY efficient, AFAIK there are still major issues with color rendering--if there weren't, we'd be replacing regular lighting with them left and right. There's also the cost problem, particularly with the looming major shortage of gallium and indium. I think the only reason people don't talk about it more on laptops is because they're considered to have such a limited life expectancy since they're not expandable; now that desktop monitors are becoming more ubiquitous, I think we'll be seeing more talk about it.

3. SSDs: Their disadvantages have been talked to death here and elsewhere.

Mike

Re:Some things conveniently left out (0, Redundant)

modmans2ndcoming (929661) | more than 5 years ago | (#25287187)

SATA drives are designed to spin up and down.

Server and NAS drives are not designed to spin up and down, but then, those are redundant srive systems so a drive failure there is not catastrophic.

Re:Some things conveniently left out (2, Interesting)

alienw (585907) | more than 5 years ago | (#25287357)

Spinning up and down hard drives: as discussed in plenty of places, including here on /. I believe, you can dramatically reduce the life of drives when you cycle them due to mechanical wear-and-tear.

Do you have any data on this? This is one of those commonly held beliefs that has absolutely no facts behind it. I've seen a google whitepaper that pretty conclusively debunked commonly held assumptions that drives fail because of temperature and "wear and tear". From a mechanical standpoint, this belief also does not make any sense. The only wear components in a hard drive are bearings on the head and spindle. Spinning down the drive should prolong their life, rather than shortening it.

Re:Some things conveniently left out (0)

Anonymous Coward | more than 5 years ago | (#25287637)

And while LED backlighting is VERY efficient, AFAIK there are still major issues with color rendering--if there weren't, we'd be replacing regular lighting with them left and right.

RGB-LED backlights have the best color reproduction ability to date. It's what the best displays for contract proof colors use. The gamut is wider than with any other display technology and the warm up time is very short. You look directly at the light source and if the colors of the LEDs are selected to stimulate the receptors in your eyes with the least overlap, then the display can create much purer color impressions than a display with a less spiky backlight spectrum.

LED lighting is a whole different story, because you don't usually look directly at the light source but at surfaces which the it illuminates. The light which enters your eye is a convolution of the output spectrum of the light and the reflectance spectrum of the surface. These are both continuous and "accurate" lighting must therefore closely resemble the spectrum of a "black body radiator."

Savings ~= Storing (1)

ZeroNullVoid (886675) | more than 5 years ago | (#25287109)

Why is using batteries, capacitors, or memory cells not considered power savers?

Isn't that their purpose...

(ignore the technicality of how most batteries are just an reaction generating power vs saving power)

Great Example of Datacenter Power Management (1)

autocracy (192714) | more than 5 years ago | (#25287133)

Microsoft's Windows Messenger (MSN Messenger, Live Messenger... whatever they call it these days) Group wrote an awesome abstract [microsoft.com] of how they cycle servers on & off to handle the load while saving power.

Now, for reasons pointed out in other comments, TFA from Infoworld is a mix of good info and horseshit.

I'd Get Fired If I Followed These (2, Insightful)

mpapet (761907) | more than 5 years ago | (#25287173)

Myth #2 suggests making your customers wait. That might work in super-mega-corporate land where your customers are literally married to you and queues in Tech Support are "profitable."

I would *deserve* to be fired if I made a customer wait. Of course, that sense of urgency doesn't work in super-mega-corporate entities either.

The myth about going to DC to be more efficient is painful too. If a manager in a workplace would entertain a crackpot ideas like that, I'd leave.

Re:I'd Get Fired If I Followed These (0)

Anonymous Coward | more than 5 years ago | (#25287291)

It would be more efficient, but might not be worth the expense.

Re:I'd Get Fired If I Followed These (1)

spidey3 (570347) | more than 5 years ago | (#25287647)

It is also not viable for time-critical applications. We lose customers when latency goes higher than 10 milliseconds! Waiting seconds for more capacity to come online is totally a non-starter for us.

What do they mean by "suspend"? (1)

argent (18001) | more than 5 years ago | (#25287257)

Suspend saves the state of the system to hard disk, which reduces the boot time greatly and allows the system to be shut down. Sleeping continues to draw a small amount of power, between 1 and 3 watts, even though the system appears to be inactive. By comparison, Suspend draws less than 1 watt. Even over the course of a year, this difference is probably negligible.

I though "suspend to disk" and "hibernate" were synonyms. A suspended computer shouldn't draw any more power than a computer that's turned off, because a suspended computer is off: you can pull AC and battery, ship it to Patagonia, plug it back in again, boot it from the hibernate image, and it shouldn't be able to tell it hadn't been on power the whole time.

Or is there some third level of "sleep" they're talking about here?

Debunking the debunkers (0)

Anonymous Coward | more than 5 years ago | (#25287275)

1. Powering a computer or server up and down limits its life span.

True. Computers contain mechanical components which are subjected to increased stress on power up. Whether that limits the life span so much that the computer fails before it has become obsolete is another question. Fact: Most electronics die on power up.

2. It takes too long to cold-start servers to react to spikes in demand. If customers are made to wait, they'll go elsewhere.

True. Not only will you annoy your users if they have to wait for more servers to be brought up, you'll also need more servers to handle the severe spike when the waiting customers all want to access the servers at the same time. If you expect spikes, have the hardware ready (or even out the load in a virtualized environment.)

3. The power rating (in watts) of a CPU is a simple measurement of the system's efficiency.

Well, there's nothing to debunk here. That's just stupid.

4. It's better to pack one big server with all the RAM, CPUs, and peripherals it can hold rather than to use multiple smaller servers.

It usually is, especially if you use components with low idle power consumption. Unless you really oversize it, the power savings from using fewer PSUs alone is worth it.

5. LCD monitors use a trivial amount of power, so you might as well leave them on. Their colors and backlight brightness improve with warm-up time.

LCD monitors certainly use non-negligible amounts of power, but their colors do change slightly during warm-up. The power savings from turning a display off do not outweigh the cost of a person waiting for the display to warm up.

6. A notebook doesn't use any power when it's suspended or sleeping. USB devices charge from the notebook's AC adapter.

The terminology is sloppy. Suspend to disk (hibernation) uses exactly as much energy as Off. Sleep ("Suspend to RAM") uses a little more. USB devices to charge from the USB port which is indeed powered by the notebook's AC adapter. Cheap PSU bricks use power even when the device doesn't draw any.

7. Notebook batteries just wear out. There's not much you can do to make them last longer.

"Many laptops with nickel-cadmium batteries.." Please! There is no such thing. You'd be hard-pressed to find a notebook with NiMH batteries. LiIon batteries do just wear out, but the rate depends on how they're used. Even under best usage conditions, they don't last much longer than 3 years.

8. Flash SSDs (solid-state drives) reduce the amount of power consumed by a laptop.

True. It does not depend on the application either. All benchmarks which have found different results where methodically flawed.

9. Going to DC power will inevitably save energy.

That's just stupid. Nobody believes that you don't need to know what you're doing when it comes to power supply engineering.

10. You're bound to save money by rushing out and buying the most energy-efficient equipment as soon as possible.

That's just stupid, period.

Re:Debunking the debunkers (1)

Penguinoflight (517245) | more than 5 years ago | (#25287473)

Don't apply for a job at Infoworld, your knowlege level is too high.

Re:Debunking the debunkers (0)

Anonymous Coward | more than 5 years ago | (#25287587)

You might want to double check your battery info:
Nickel-cadmium = NiCd, NiMH = Nickle Metal Hydride.

As far as laptops using them, I've got 2 old laptops that use a NiMH, and an even older one that uses a NiCd.

(By old, I mean 8, 10, and 13 years I think, about the only one I can confirm is the 8 year old one(got 2nd hand from sister) as the other 2 were bought at surplus sales )

And I have a Lithium-Ion in my most recent (and first one I got new) laptop. (Purchased about 1 month ago).

So maybe finding a new laptop using NiMH or NiCd, but they do exist.

Re:Debunking the debunkers (0)

Anonymous Coward | more than 5 years ago | (#25287853)

There is no such thing as a laptop with NiCd batteries. (It is illegal to introduce one into the market here! Talk to a museum if you find a NiCd battery powered notebook, but don't write about it in an article about energy myths.)
You'd be hard-pressed to find a notebook with NiMH batteries. (They exist but they're quite rare nowadays.)

Better?

They have their own fans... (3, Funny)

pmandryk (858235) | more than 5 years ago | (#25287391)

My favourite story (or urban legend) is when an employee came in to an IT shop on the weekend and shut down all of the A/C cooling units for the Data Centre. He claimed that he was "going 'Green' and saving power" because "...all of those computers in that room have their own fans." I'm pretty sure he was let go after that...or promoted to management.

I remember why I don't read Infoworld (3, Insightful)

sheldon (2322) | more than 5 years ago | (#25287393)

That's a really bad article. Wow, worse then anything I remember them writing before.

LED backlit LCDs? (0)

Guysmiley777 (880063) | more than 5 years ago | (#25287419)

@ #5:

LCDs with LED backlighting rather than fluorescent don't need any warm-up time at all.

Yeeeeeaaaaaahhhhh, LED backlit LCD monitors run in the $1,500+ range. Soooo if I save $10-$40 per year, I'll break even in 30 to 120 years. The mind boggles.

Bad article... (2, Informative)

Junta (36770) | more than 5 years ago | (#25287551)

Out-right potentially wrong: no one cares if a customer is made to wait for a server to boot to get served. That's not a generalization to be made lightly... It is true, though, that suspend-to-ram has not received the attention it deserves in the data center. A great deal of server-class systems and options are not designed to cope with suspend-to-ram, and thus you must be careful banking on this. The industry should correct it, but a facility can't bank on it yet (just put pressure on your vendors to make it so...)

Straw-man: A supposed 'myth' that leaving on LCD monitors is fine for energy savings, with the remarkable clarification that being off saves more power... Who would have thought.

Other straw-man: You will unconditionally save money by rapid upgrades to the latest efficient technology. I don't think anyone is foolish enough to think compulsively following any technical treadmill will lead to any overall financial gain..

Re:Bad article... (2, Informative)

Volante3192 (953645) | more than 5 years ago | (#25287619)

Well, obviously leaving LCDs on saves energy if you compare it to leaving CRTs on.

Daylight Savings Time (3, Informative)

flyingfsck (986395) | more than 5 years ago | (#25287677)

Probably the biggest and most annoying/disrupting power saving myth is Daylight Savings Time. Every year, the power companies announce that they don't notice any change whatsoever in power consumption.

High efficiency power supplies (1)

disordr (1069522) | more than 5 years ago | (#25287815)

In my latest batch order of servers (20 x 1U machines) I've got a config made up of: - intel xeon quadcore X3320 @ 2.5Ghz. These are the "efficient intel cores) - 4 x 2GB unbuffered ECC memory - 3ware 9690SA in RAID-10 with - 4 x Seagate SATA 500GB es.2 drives - high efficiency 280Watt PS while I haven't plugged them directly into a kilowatt meter, the first one I plugged into my metered APC unit showed it using .7amps at idle. Compared with some large honking 2U's with about double the power (2 x quadcores, 16gigs memory, 6 disks), these 1U systems use less than half the power rating as those larger machines. Overall this has been a big win in datacenter cost savings for our 20amp circuits which we get charged for monthly. I can get a bunch more machines on a single circuit.
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>