S3 Standby State Done Right

For Earth Day, Cameron Butterfield has written in with a pointer to his article on how to get your Windows PC into S3 sleep, and why you want to. It covers the question of how to take advantage of this extremely low-power mode even when your machine is an "always on" file server, remote desktop, or VNC server.

And Linux?

[Ed Avis]

Great for Windows users... but what are the options to set up a Linux system to reduce power usage and fan noise when idle?

Re:And Linux?

[BACbKA]

Gentoo's Power Management Guide [gentoo.org] is a bit gentoo-centric, but most things carry to another distribution easily.

Re:And Linux?

[rduke15]

See this article: Debian HOW-TO : CPU power management [technowizah.com]. I used the info to configure a couple of Poweredge 860 server. Most of the time, it's at a CPU speed of 300Mhz instead of 3 Ghz. That saves quite some power, and you cannot notice the difference in speed.

Blame Bill Gates if it does not work.

[twitter]

Anyone who has not Bill Gate's memo about this [edge-op.org] should. Anything M$ touches is shit: winmodems, wifi, ACPI, APM and the list goes on and on. They can't make their own stuff work, so they have to break everyone else's.

Despite his efforts, power management can be made to work. It's not easy and you can't expect the latest and greatest to work. The closer a company's working relationship to M$ is, the harder it will be to make things work. For example, Dell is more difficult and Thinkpad is easier. As

Nice FUD

[The Bungi]

Anything M$ touches is shit

Oh yeah.

Bill Gate's memo

That's an interesting email from 1999. Myself, I've been known to send emails to the tone of "how can we prevent the competition from leeching on our multi-million dollar R&D investment with our technology partners", but OK.

Would you like to point me to the follow up email from Eric Rudder that says "Hi Bill - As you requested, we've made the ACPI extensions specific to Windows so no one else can implement them. Cheers!" I can't seem to find it.

Oh, wait - here's ACPIfor Linux [sourceforge.net] and ACPI for FreeBSD [freebsd.org]. Indeed, here's a quote from the WP entry:

The Advanced Configuration and Power Interface (ACPI) specification is an open industry standard first released in December 1996 developed by HP, Intel, Microsoft, Phoenix and Toshiba that defines common interfaces for hardware recognition, motherboard and device configuration and power management.

Now, ACPI has its shortcomings. It's complicated. It might not be your ideal of a standard. But it is an open standard, which Linux indeed implements. It might be broken in some ways in Linux as it is in Windows, but implemented it is. It's an important standard because it takes hardware out of the equation, which is important for a general OS that's supposed to support a wide range of it.

I still use APM for the most part

Really? That's also a Microsoft-defined standard [wikipedia.org] (along with Intel):

Advanced Power Management (APM) is an API developed by Intel and Microsoft

Is that standard "shit" as well? And if you all these standards from Microsoft are "shit", then why do you use them at all? You use Linux, right? Why don't you come up with your own standard and give it to the free software world so they can stop using all these "shit" open standards that Microsoft has bothered to make open for anyone to use? Which reminds me, I'd love to see that other email about ACPI I mentioned. Thanks.

Re:

[xenocide2]

Unfortunately, the world is far more complicated than you'd like. Matthew Garrett has [livejournal.com] extensively [advogato.org] about the subject. Truth is, ACPI a standard that nobody follows intelligently. Garrett writes about how part of the spec involves an interpreted machine code called DSDT (this already sounds like a recipe for disaster) that is used to guide actions. The problem is two fold:

• DSDT's are buggy (go figure)
• The common method for fixing a broken DSDT is to patch it after the machine has booted via some driver

Micros

Re:

[xenocide2]

That post looks a bit screwed up that I made, but he's a central member of Ubuntu, a member of the laptop, kernel, and acpi teams and one of 4 members of the technical board. I had hoped his insightful analysis would have been enough, but it seems I botched a link or two. To make up for it, here's a video [linux.org.au] of him detailing how hacking acpi is done.

Of course it does

[The Bungi]

LinuxBIOS or other free software implementations are much better.

Yes, that's been a great success. Are you working on this or other standard to replace all the "shit" Microsoft has given you so far? How's that going?

I'm surprised Bill was dumb enough to document his intent.

What you are implying is action. I asked for you to document how Microsoft "screwed" Linux power management with ACPI, possibly screwing themselves (if I follow your logic) in the process, since Windows contains an ACPI implementati

Re:APM Sucks too.

[The Bungi]

Oh twitter, you're going to love this. Here's an article [advogato.org] by the guy that actually writes this stuff for your operating system. I'd like you to go through that article and please share with us where the guy that actually writes the stuff blames Bill Gates or "M$" for how ACPI works (or not). As opposed to just the general sense of "this stuff is hard" I get from it.

Once you're done getting an education, I'd like for you to explain how "M$" allegedly sabotaged ACPI on Linux. You pointed [slashdot.org] to an eight-year old email from Bill Gates that, if anything, proves Microsoft did not do anything to impact the implementation of ACPI in Linux. Seriously, just in case your FSF distortion field is turned up too high, that's exactly what you are proving by linking to that email. You have ACPI in Linux. It might be as broken as it is on Windows, but you have it. You realize that, yes? God, please tell me you realize that?

Re:

[Achromatic1978]

Generally people don't bother to refute the content of his posts because it's pointless. He'll argue until he's blue in the face - or you provide a coherent and logical response to him. He has no rebuttal for that. Far easier for him to go on to the "Evil Micro$oft WinDoze!" troll.

Re:

[bcrowell]

what are the options to set up a Linux system to reduce power usage and fan noise when idle?
I just built an amd x64 dual-core system, and it only draws 51 W for the whole system when the screen powers down. AMD has something called cool'n'quiet, which was supported automagically when I installed Ubuntu Edgy. It ramps down the cpu frequency when the processors are idle. Because the system draws so little power, I was able to disable the case fan, so now all I have running is the cpu fan and the power supp

Re:And Linux?

[BillGatesLoveChild]

> what are the options to set up a Linux system to reduce power usage and fan noise when idle?

Disconnect those pesky cooling fans. They just make a lot noise and suck up power. Truth is, your PC will run fine without them. It's just a scam by equipment manufacturers to make a few extra bucks out of you. I've been running with them removed for years, no problems.

regards
Scott E. Brown
NOAA Antarctic Station

Re:

[Sigma 7]

Disconnect those pesky cooling fans. They just make a lot noise and suck up power. Truth is, your PC will run fine without them.

"You've got a hole in your mainboard. [google.com]"

On a more serious note, cooling fans are recommended as they help keep the CPU cool. PC case fans are considered optional, but can be used if your CPU is reporting temperatures that are considered higher than what they should be (which may actually be ambient heat from the power supply.)

When referring to laptops or notebooks, it depends on the model. While my notebook is at a normal temperature most of the time, I have been running a few 100% CPU tasks which does hea

Re:And Linux?

[Thomas Shaddack]

A way to make a computer quieter is replacing the smaller fans with somewhat bigger ones, and slow down their rotation (eg. with a suitable series resistor). The aim is to get comparable airflow over the heatsink with lower fan blade speed, which means less turbulence over them, which means much less noise.

Re:

[flyingfsck]

Here you go:
service syslog stop
hdparm -B 1 /dev/sda
hdparm -S 5 /dev/sda
echo 5 /proc/sys/vm/laptop_mode

Re:

[smorar]

I found an awesome blog post the other day which explained how to get a desktop linux box into S3 suspend. It's all about your BIOS settings. Now, I can turn my pc 'off', and it takes less than 5s to resume. http://shallowsky.com/blog/linux/desktop-suspend.h tml [shallowsky.com]

Re:

[Asztal_]

You realise that would be finished in 15 seconds and then you could put it to sleep, right?

Also available on Linux

[Anonymous Coward]

It doesn't seem to be a hot topic because I couldn't google a definitive page. There were lots of pages for individual computers or distros though.

The documentation is probably on your own computer at: /usr/src/linux/Documentation/power/ ... The exact file on my system is states.txt but it also seems to exist on other distros as suspend.txt

Laptops?

[Vo1t]

I wonder if how does S3 work on a laptop? Does laptops' built-in energy saving mechanism collide with tricks described in the article?

Re:

[Southpaw018]

Most laptops come preconfigured to take advantage of most of the stuff in the article, though it wouldn't hurt to check. The last few new Dell laptops we've purchased at my organization default to S1 after a few minutes, S3 if you close the lid or hit sleep, and S4/S5 for shutdown.

S3 and MCE

[gEvil (beta)]

I use the S3 standby on my MCE machine, and it's really really nice. I turn the machine on and off (well, awake/asleep) using the power button on the remote, and the machine is up and ready to go in about 3 or 4 seconds (as long as it takes me to switch the TV to the right input). I've only ever had it refuse to wake up once in the 1-1/2 years I've been using it, and that was remedied by using the power button on the front of the machine (it woke right up and didn't even need to be rebooted). Definitely worth looking into for instantaneous access + decent power savings.

Re:S3 and MCE

[MSRedfox]

Under MCE, I use the MCE Standy Tool. MCE has a bad habit of waking up to record a show and then not returning to standby afterwards. This can result in the computer staying on all day instead of just 1/2 hour to record a small show. The Standby Tool has features to help MCE handle power management in better ways then Windows default methods. It makes me wonder why Microsoft couldn't get things to work as smoothly as this 3rd party software. http://www.xs4all.nl/~hveijk/mst/indexe.htm [xs4all.nl]

Re:

[CommanderData]

Thanks for the link, I never tried to look for a solution... My MCE will have bouts of trouble where it will not return to standby after overnight guide updates. It seems to "cure" itself for a week or two before starting the same trouble again. Now I don't have to bother writing my own standby utility :)

Re:

[MSRedfox]

It happened after watching live TV for me. If I was watching a recording or DVD and suspended, it would re-suspend after recording new shows just fine. But if I pressed the PC suspend button on my remote after watching live TV, the next time it woke up to record a show, the PC would stay on after the recording was done. Even if I pressed the pause or stop button before suspending it would still not power back off correctly. For a while there, I would always start playing a recording before suspending th

network broadcast traffic

[Anonymous Coward]

Surely enabling your PC to wake up whenever any network traffic is sensed is stupid in the example described in the article.

Will it not wake up whenever any workgroup broadcasts are sent to it?

Re:network broadcast traffic

[Bozzio]

yes it will.
Or it does for me. Even if the computer is alone on the router. It seems my router occasionally broadcasts something and wakes up all my computers.

I've switched to using the magic packet alternative. The only problem is that since my server PC is behind my router, I have to SSH into the router and sent the magic packet from there. ICKY.

I hear other routers (mine is a Linksys WRT54GS) will let you WOL remotely. Normally, you just send your magic packet to the router and set up the router to convert it to a broadcast.

If I remember correctly, a magic packet is just a packet with the correct header and the client's MAC address broadcast to the network.

Re:

[SplatMan_DK]

I do not have the same problem, and I am allready using S3 powermode.

A "broadcast" package is not a package with a single specific receiver, so why would a machine in S3 mode wake up when it detects a broadcast package? The whole point is to make sure the machine only wakes up from LAN access when there is traffic directed specifically for that interface/address?

FreeBSD

[evilviper]

I know I'm setting myself up for flames around here, but the OS with the best support for APCI S3 Suspend is FreeBSD 6.2, even though it's certainly not perfect.

My desktop _almost_ worked. I had to swap-out my ATI video card to get it to resume from S3.

Now, the big problem is X.org... Since X doesn't play well with suspend, FreeBSD is supposed to switch off of X, to a virtual console before entering suspend mode. Unfortunately, I've found that, unfortunately, X 6.9.0 freezes about 1 in 3 times. Once I figured that out, it was just a matter of manually switching to a console, then typing "suspend" before I walk away. Now I haven't rebooted my machine in months, and it's on and usable (right where I left everything) in about 3 seconds.

Of course, the drawback to X not cooperating is that I can't set my machine to auto suspend when it's been idle for a few minutes, but I'm hopeful the next release of FreeBSD will fix that. X6.9.0 is the latest ported release, and compiling from vanilla sources goes horribly, horribly wrong, right now. I could try downgrading, but it's not worth the hassle and lost features, IMHO.

Re:

[Piquan]

As a FreeBSD user, I'd be interested in hearing how you configure this.

Re:FreeBSD

[Anonymous Coward]

Guys! You found each other!

Shouldn't you guys exchange phone numbers or something?

Re:FreeBSD

[evilviper]

S3 mode is entered by running "acpiconf -s 3"

All available options can be listed by running "sysctl -a hw.acpi" and included in /etc/sysctl.conf to be automatically set upon boot-up. Basically you'll only need "hw.acpi.reset_video=" set to 0 or 1 depending on your system.

If you need to unload modules or any other action before suspending, see /etc/rc.suspend. Put the opposite commands in /etc/rc.resume.

That should be everything you need. Either your hardware will work, or it won't. In the latter case, strip your system down to nothing but video, and try different video cards. Then add a piece at a time to see what's causing problems.

Re:

[Breakfast Pants]

Take a look at the command chvt, and you may be able to script it and put the script as a button on your panel/dock/whatever.

Re:

[evilviper]

Take a look at the command chvt,

I looked, and found nothing.

chvt isn't installed on FreeBSD, it isn't available in ports, Freshmeat.net doesn't know about it, a web search turned up man pages for it but no source code, etc., etc.

Re:

[evilviper]

Pretty unlikely it will even work on FreeBSD, then.

Re:

[Breakfast Pants]

Ah sorry, I thought it was a standard unix tool. I had to use it on my laptop running linux to get the screen to come back up after it was turned off by power management (the only way to get it to come back on was to switch to a VT and then back to X, I wrote a script to run chvt to do this and had it triggered by xlockmore and it worked like a charm).

Re:FreeBSD

[value_added]

I know I'm setting myself up for flames around here, but the OS with the best support for APCI S3 Suspend is FreeBSD 6.2, even though it's certainly not perfect.

Perhaps, but the issue is a lot more complicated than that. We're talking about the BIOS, the OS, and then how the two relate to each other. That said, it doesn't suprise me that the article is lame. Setting a fixed IP address and making use of WOL? What's that got to do with Windows, and what does "done right" refer to?

The only informative (and amusing) bit was the Microsoft chosen USB behaviour (hidden) that requires an "easy" registry edit to change. So much for "Oh, no, not manually editing a config file!" I guess having all the behaviour and options explicitly set forth and easily editable is the wrong approach. ;-)

X.org... Since X doesn't play well with suspend, FreeBSD is supposed to switch off of X, to a virtual console before entering suspend mode. Unfortunately, I've found that, unfortunately, X 6.9.0 freezes about 1 in 3 times. Once I figured that out, it was just a matter of manually switching to a console ...

I'm going by memory here, but IIRC, that's handled with a sysctl. You shouldn't need to manually do anything. Read through acpi(4) and then Google for more info, or better yet, just search the 'mobile' archives for some possible settings and the merits of each.

Of course, the drawback to X not cooperating is that I can't set my machine to auto suspend when it's been idle for a few minutes, but I'm hopeful the next release of FreeBSD will fix that.

I'm not sure you want an S3 state every few minutes. It would make more sense to blank the screen (and kill the backlight on a notebook) by setting the DPMS option in xorg.conf, and set your screensaver options in .Xdefaults. The CPU can be trottled using any number of methods either on a dynamic basis, or at set time. Throw ataidle into the mix and you've got most everything you need for those "every few minutes' intervals. How many more options could you want?

For a full suspend after x minutes, why not script your own approach? One option would be to use xscreensaver-command to invoke a count-down timer to invoke zzz(8)? Or if power usage is a Really Big Deal, make use of WOL and start/stop the system at set times. Dunno if that would work for a desktop system, but it might cut down the hours on /.

Re:

[evilviper]

Perhaps, but the issue is a lot more complicated than that. We're talking about the BIOS, the OS, and then how the two relate to each other.

It's certainly complicated, but FreeBSD seems to have the most people putting in effort to get it working, working-around bugs in the hardware, etc.

I'm going by memory here, but IIRC, that's handled with a sysctl. You shouldn't need to manually do anything.

First, there's no sysctl. Second, yes, it is supposed to handle it automatically, but it's buggy, so manually doin

Re:

[value_added]

First, there's no sysctl. Second, yes, it is supposed to handle it automatically, but it's buggy, so manually doing it is the only way to be sure. Though, I could have sworn I said that the first time...

So what exactly does 'sysctl -d hw.acpi.reset_video' return on your system?. Or 'sysclt -d hw.syscons.sc_no_suspend_vtswitch'?

I don't understand the emotional aspect of your comments, but FWIW, yes, there could be improvements to default values, default configurations, but that's another issue that we could

Re:

[Guy Harris]

Any filesystem which requies a disk write every few seconds on an idle system (i.e nothing is reading from or writing to the disk) is broken.

And the only reason why reading from a file system should require a write to the disk(s) it's on would be to update the last access time, so, if mount or other options are available to disable access time updates, that might help, although

1. some file systems might decide not to treat those updates as sufficiently important to push them to disk until the file system is

dumppo.exe the Microsoft Power Tool

[MSRedfox]

Windows XP will often times not give s3 suspend as an option even when turned on in BIOS. But with Microsofts dumppo.exe utility you can force it to use an S3 or S4 state. ftp://ftp.microsoft.com/products/Oemtest/v1.1/WOST est/Tools/Acpi/dumppo.exe [microsoft.com] To force it to S3, run this under command prompt "dumppo admin minsleep=s3"

Par for the couse. M$ booby trap.

[twitter]

Windows XP will often times not give s3 suspend as an option even when turned on in BIOS. But with Microsofts dumppo.exe utility you can

How typical, a DOS only power tool to manipulate your hardware and everyone else is out of luck. Yeah, that stinks. [edge-op.org] Thanks, Bill.

Re:dumppo.exe the Microsoft Power Tool

[ergo98]

I wrote about the power consumption of S1 versus S3 sleep [yafla.com], and as you mentioned dumppo.exe was the enabling-tool that let me take advantage of this great bit of functionality.

The biggest downside of S3 sleep is that about 1 out of every 200 recovers or thereabouts it completely fails to come back, thought that's probably a mainboard issue more than an OS or technology issue.

Oh, and a great little helper app if you use S3 is WakeUpOnStandBy [dennisbabkin.com]. It allows you to configure a machine to "come alive" at scheduled times, even from an S3 sleep (apparently the BIOS supports configured wake-up times, and this app knows to tell it to wake up accordingly just as going to sleep). Very helpful little app -- I have my PC set to come alive in the morning when I know I'll be remoting in.

Oh, and rather than waking up on all network traffic, as the article recommends, it's far better to wake up on WakeOnLan packets. There are lots of resources out there for that.

want a "file transfer" powerdown mode.

[plasmacutter]

often, my computers cant be put to sleep because theyre transferring files (over aim, bit torrent, you name it.. every app according to its need).

Ive noticed all companies, including apple, whose products i use, are giving you only a black and white choice. you either have the computer awake or its fully asleep.

i'd like to have a low power transfer mode, where the cpu is reduced (to 1 core at say 500 mhz), the monitor is turned off, and as much memory as possible is dedicated to the apps which are doing intensive file reads/writes. this will allow the hard drives to be used less by caching the files in ram and pulsing the hard disk.

Re:

[Guy Harris]

Ive noticed all companies, including apple, whose products i use, are giving you only a black and white choice. you either have the computer awake or its fully asleep.

i'd like to have a low power transfer mode, where the cpu is reduced (to 1 core at say 500 mhz), the monitor is turned off, and as much memory as possible is dedicated to the apps which are doing intensive file reads/writes. this will allow the hard drives to be used less by caching the files in ram and pulsing the hard disk.

That is "awake",

Re:

[gbjbaanb]

Well, for windows at least, if you have a "power-aware" CPU (eg the Core2Duo or AMD64) then it will automatically step itself down. My AMD x2 steps down to 1Ghz and also (more importantly) reduces the voltage used to 1.075v.

You can set the monitor to switch off anytime - the 2 'mini' powersave options are 'turn off monitor' and 'turn off hard disks' - so go ahead and put some minutes in the boxes.

You don't really want to 'pulse' the hard disks, I think that'll just wear them out. Its also better to keep the

Re:

[gbjbaanb]

Obviously, it takes more energy to spin a disk all week than it does to power it down for a read on monday morning and a write on friday afternoon. I don't know exactly at what point is it more cost effective? I generally use 1hr in the power save box before the drive is spun down. However, the OP wanted to 'pulse' the HDD, basically keeping it powered down until it was needed which would be terrible for energy use.

Bear in mind that big, fancy SCSI arrays on big boxes have special circuitry to bring the dri

Re:

[evilviper]

i'd like to have a low power transfer mode, where the cpu is reduced (to 1 core at say 500 mhz),

Thanks to AMD's CnQ, and Intel later following suit, any CPU made in at least the past year (and more for AMD64 CPUs), will idle down to low power states automatically.

the monitor is turned off,

Also easy. You can hit the power button on the monitor, you can wait for it to automatically shut off after 15-20 minutes, or with X11, you can run xset and tell the video card o shut-off the monitor.

and as much memory as

Re:

[ivan256]

You need to write to the disk every few seconds, to maintain a consistent state, with or without a journaling file system.

No you don't, and in fact if you mount your filesystem read-only, or noatime, and run noflushd your hard drives can spin down indefinitely as long as your dataset fits in memory. I used to get 8-9 hours out of the battery on my PowerBook G3 using this method and low screen brightness.

Of course, if you are writing to files and you do this and then lose power, you lose data... But you c

Re:

[evilviper]

No you don't, and in fact if you mount your filesystem read-only, or noatime, and run noflushd your hard drives can spin down indefinitely as long as your dataset fits in memory.

Perhaps noflushd will work, however, I don't know any details about how it works, and you may well risk corrupting your filesystem. Just mounting a fs noatime certainly won't work, and yes, read-only filesystems can spin-down, but that's not the situation we're talking about here.

Re:

[evilviper]

On second though, I just looked-up noflushd, and it seems it also absolutely won't help at all:

Journaling filesystems like ext3, reiserfs or xfs bypass the kernel's delayed write mechanisms. This amounts to lousy spindown times when working off such a partition. There's no workaround for this.
http://noflushd.sf.net/ [sf.net]

Re:

[ivan256]

Easy workaround: Don't use a journaling filesystem.

There is no reason a filesystem needs periodic activity in order to prevent corruption; after all, it doesn't get corrupted when your system is powered off... Journaling protects against corruption when a transaction that is in progress is interrupted. If you don't mind integrity checks after a power loss, you can work just fine without a journaled filesystem. People have done it for decades. We can go with your theory that you can't do this, or we can go w

Re:

[evilviper]

Easy workaround: Don't use a journaling filesystem.

On Linux, that's not easy at all, and noflushd appears to be Linux-only.

The only reasonably capable, non-journaled file system is Ext2. Using Ext2, you have the choice of either mounting it read-only, and getting horrible performance, OR leaving it async (default) and running an extremely high risk of corrupting your entire file system in the event of power outage, or other system crash... Note: that means corrupting your ENTIRE file system, not just losi

Re:

[ivan256]

It's a calculated risk. It's the same calculated risk that battery backed-up RAID controllers take (Will the battery die first, or will the disks become available to persist that data to?), except that they do it for performance, not power savings. It can be reliable s you're willing to spend power on, and the performance is very good. With a little dicipline (or a well written script), you shouldn't have to worry about ever losing more than 15 minutes of data, and you should never end up with a corrupted f

Re:

[evilviper]

you should never end up with a corrupted filesystem.

Hate to tell you, but that's exactly what happens. You can't guaranty file system consistency when doing out-of-order writes. That's among the main reasons Ext3 does a full "fsync" every 6 seconds. It's a serious limitation of Ext2 in it's default mode. You're simply playing Russian Roulette with your data, and eventually you'll lose.

You certainly don't have to take my word for it. It's a well known issue, and there are several write-ups of it.

Re:

[dbIII]

Note: that means corrupting your ENTIRE file system, not just losing the file you're downloading

Do you mean only corrupting to the point where you have to use fsck or something unrecoverable without more work? How is this worse than other non-journaled file systems like ufs which are also fixed by fsck after unexpected power loss?

Re:

[evilviper]

Do you mean only corrupting to the point where you have to use fsck

No, I mean CORRUPT. As in data loss. Seemed a pretty simple term.

Re:

[Chandon Seldon]

an extremely high risk of corrupting your entire file system in the event of power outage

I've never heard of an ext2 file system not surviving a power failure. Sure, it has to be fdisked. Sure, sometimes a file or even a directory is corrupted (in the latter case stuff shows up in "lost+found"), but the file system as a whole is usually fine.

Re:

[ivan256]

noatime prevents reads from making journal transactions which spin up the disk. If you're serving static files (I.E. Reading only), you can mount noatime and prevent the disk from spinning up once all the data you're serving is in memory.

Another old trick, which I still use regularly, is to copy all the data you're serving into a ramdisk or a tmpfs, and then unmount all disk based partitions. Turn on PowerNOW! or SpeedStep, force the CPU multiplier low, and you can serve thousands of pages per second for un

Re:

[evilviper]

noatime prevents reads from making journal transactions which spin up the disk.

No, it doesn't affect the journal activity at all. atime is absolutely not the main problem with modern file systems spinning down.

They need to constantly update the journal, or other file system info, atime or no, every few seconds to assure whatever fsck program, that the fs info contained in the journal or elsewhere is up-to-date. Of course that COULD be worked around, but that's how modern file systems work, and it's certai

Re:

[Guy Harris]

They need to constantly update the journal, or other file system info, atime or no, every few seconds to assure whatever fsck program, that the fs info contained in the journal or elsewhere is up-to-date.

They need to update the journal, or otherwise push stuff to disk, if there's stuff on disk that's not up-to-date with what's in memory. Once that's done, everything that was pushed from memory to disk is now up-to-date on disk, so only stuff changed in memory after it was pushed to disk needs to go to dis

Re:

[Graff]

plasmacutter had this to say:

Ive noticed all companies, including apple, whose products i use, are giving you only a black and white choice. you either have the computer awake or its fully asleep.

i'd like to have a low power transfer mode, where the cpu is reduced (to 1 core at say 500 mhz), the monitor is turned off, and as much memory as possible is dedicated to the apps which are doing intensive file reads/writes. this will allow the hard drives to be used less by caching the files in ram and pulsing th

Re:

[Guy Harris]

thats not what i observe from my g5.

That's not a modern Macintosh.

I don't know which of the PPC processors used in Macs supported that sort of speed adjustment.

Re:

[Guy Harris]

total bunk, that was only 3 powermac revisions ago.

OK, it's not a sufficiently modern Macintosh. The PowerMac G5 family is almost 4 years old. The 970FX supports frequency scaling (see the IBM PowerPC 970FX RISC Microprocessor User's Manual [ibm.com]), but I can't find any user's manual for the earlier 970's, so I don't know whether they support it as well. If they don't, you're out of luck, and need to get a more modern machine.

I also don't know whether it would be possible to switch off individual processors i

Re:

[Guy Harris]

your condescension and use of the word "modern enough" is getting annoying. core duo chips are less than 18 months old, so enough with the snobbery.

A PowerMac G5 with a 970FX is a more modern machine than one with the original 970, and this "New Product Focus" note [ibm.com] seems to indicate that the 970FX was the first 970 to support changing the clock speed and voltage on the fly, so, if you want to have the OS able to adjust the processor speed (or the processor able to adjust its speed spontaneously), and you h

Slow down HD?

[WoTG]

I've been wondering if it's technically possible for newer 10K or 7.2K HDD's to slow down during quiet periods. It bugs me that the drive in my little home server has to run 24/7 at full speed just to receive the odd email or web page hit every 5-10 minutes.

Maybe there isn't a big enough power difference to make it worthwhile, I dunno.

Anyway, with flash is getting so cheap, sooner or later I'll find a way to delegate those things to flash and let the HDD actually go into sleep mode at night.

Re:

[QuasiEvil]

A laptop drive takes about 1.5W just sitting there, whereas a 7200rpm desktop drive usually idles at about 8-15W, depending on the breed. My house server got its main drive replaced by a 5400rpm laptop drive about a year ago. Much quieter, cooler, and lower power, and there's no real downside since nothing on there particularly requires high speed disk access.

Re:

[springbox]

AMD's processors feature the so called "Cool and Quiet" technology. I have an Athlon 64 and it works great. The CPU's frequency and voltage is reduced when it's not being used for much, which is really quite often. It switches back to normal (maximum frequency) when the processor becomes loaded, along with the fan, which only spins faster when the CPU starts to become a bit warm. It's not ideal but it's better than running at full blast all of the time. Some of NVidia's graphics cards will also do the same

thanks!

[hansamurai]

Wow, great article and definitely something I wouldn't even consider unless it discussed wake-on-lan settings as I use my computer as data storage for my media center also. I tested out my standby settings and my fans just kept going, which is a problem for me because right now my office is 9 degrees hotter than the temperature outside (80 to 71, in Minnesota!). Kind of uncomfortable. Also nice to see an article all on one page. I expect to see a regurgitation of this article soon on some ad-ridden PC s

Re:

[Doppler00]

Yeah, I think this is a huge problem with power management. Unless you use the fan connectors on the motherboard, the cooling fans keep going. Would be nice if the power supply could add some intelligence and make specific fan connectors that shutoff when sleeping.

What about wireless ?

[Qwavel]

There was some useful info in this article about configuring your network adapter to support wake-on-lan, but what about wireless adapters? In my experience they don't seem to support WOL or any equivalent. The only solution I can think of is to connect an ethernet client device to my computer so that I can use the WOL of the computer's ethernet, but this is not really a good solution.

Is there any sort of WOL capabilities in the new 802.11n?

Re:

[Guy Harris]

There was some useful info in this article about configuring your network adapter to support wake-on-lan, but what about wireless adapters? In my experience they don't seem to support WOL or any equivalent.

At least at one point, I found one 802.11 adapter or chipset that supported OnNow-style [microsoft.com] wakeup, but I don't know whether drivers supported that.

You'd have to keep the radio on, though, which means there's some power you can't save.

Is there any sort of WOL capabilities in the new 802.11n?

That's probably

Bad Assumption

[Anonymous Coward]

* I calculated (24 hours per day) * 30 days a week [sic] = 720 hours
* Power bills are generally measured in kilowatt hours or "kW/h"s. Power rates might be as much as $0.12 per kW/h
* Our total cost of having the computer on 24/7 for the month in this scenario would be as follows:
* .4 kW (400watts) * 720 Hours * $0.12kW/h = $34.56

The 400 watts per hour is a really poor assumption. First, the average home PC wouldn't use 400 watts at peak, let alone continuously. Secondly, the 400 watts would almost never be continuous. Even a PC left on overnight will use far less power then one being actively used during the day. My power costs almost $0.11 per kilowatt-hour, and I have a power bill during non-summer months (damn AC) of about $45-$55 dollars with two desktop computers running 24/7. I honestly wouldn't be surprised if most of this came from my computers, but it certainly isn't as much as the article makes it sound.

That said, it is a good article on how to keep the "instant-on" without using excess power.

Re:

[TheThiefMaster]

It depends on the power supply. The cheap ones can be as bad as 50% efficient, in which case a 200W pc would draw 400W from the wall.

Incidentally I have two PCs (with >85% efficiency PSUs) and a 19" CRT monitor plugged through a power-meter right now and they are drawing 510W total, and 425W if I turn off the monitor.

One of the PCs is a dual socket A machine with cpus that won't go below 60C despite some really powerful air cooling, and the other is an AMD A64-X2 3800+ with ATI X1900XTX. Both are fairly

Calculations are a bit off

[bogie]

"If we take just a reasonable estimate that a computer uses 400 Watts idling along, we can find some astounding figures."

That doesn't sound very reasonable to me.

".4 kW (400watts) * 720 Hours * $0.12kW/h = $34.56"

Nope, that's way off what the average PC costs to run.

He does have a point thought about using lower power modes. On newer PCs it seems to work well and it will save you bucks if you have several PCs in your house.

Re:

[pimterry]

I have a power monitor thing on the socket for my home server (it's just a box, no screen keyboard etc) and right now it's using 132 Watts downloading torrents and web serving (mostly as a web dev test site, so probably not really doing any work). It's a 3Ghz P4 too, so it's probably not as power efficient as it could be.

400 watts has got to be way off.

400 watts is high

[jbengt]

I work as an HVAC engineer, and I have to take into account the PCs when designing air conditioning for an office. I figure 200 to 250 watts per workstation; that is supposed to take into account average usage including everything: the PC, monitor, peripherals, task lighting, occassional printers, etc. I've been told that this is too high, but my career has spanned a lot of changes - dummy terminals, energy inefficient monitors, heavy duty PC workstations, efficient but larger monitors, LCD monitors, thin

Re:Calculations are a bit off

[Michael Woodhams]

Those 600+ W power supplies are purely for people with inferiority complexes about other aspects of their lives/bodies. Here's [silentpcreview.com] a discussion about how much you can run on a 300W PSU. 300W suffices for a modern high-end CPU plus high-end GPU plus a bunch of drives, when under heavy load. Even a high end system will idle at around 150W. A more sensible system is probably idling around 80W.

NOTE: All the figures above are *not* including losses in the PSU. A modern PSU should be about 7 5% efficient, so increase the above by 1/3 to make them comparable to the 400W number in the article.

Re:

[QuasiEvil]

Finally, somebody said it. I completely agree with you, and something else that should be taken into account is the fact that the more closely matched your power supply design capacity and your load, typically the more efficiently the supply will run.

There is *one* setup that could require the BFPS - Geforce 8800 GTXs running in SLI. Those bastards chew up like 160W each. Figure 360W total just for the two vid boards. Then figure another 150W to run the rest of the system at peak load (drives, dual core

Re:

[Michael Woodhams]

There is a problem that nobody seems to be interested in producing good low capacity PSUs. If you want (say) 80% efficiency and modular cabling, most manufacturers don't have anything below 500W. (A few have 400W.)

Some good options now are Seasonic S12, Antec NeoHE, Silverstone ST30NF, Nexus NX-80x0 series.

My weeks only have 7 days in them.

[QuietLagoon]

From the article: I calculated (24 hours per day) * 30 days a week = 720 hours

Does that mean my PC costs one-quarter of what he calculates?

Re:

[flyingfsck]

It means that he is a consultant and bills per hour and that is what he bills per week...

Not really applicable to servers?

[dave562]

It seems to me like this is more of a workstation thing than a server thing, at least in Microsoft land (which is what the article is about). In any sort of domain environment, the DC is going to be talking to the servers at least every fifteen minutes, if not more frequently. The servers won't be asleep for very long.

Re:

[flyingfsck]

The servers may still be asleep for 14 out of every 15 minutes on average. Anyhoo, I'm not so much concerned with energy use as fan noise. I find that disabling the syslog daemon and shortening the HDD sleep timeouts to 25 seconds make an enormous difference in the noise level. Laptop mode on Linux will also speed step the CPU with an accompanying reduction in fan noise.

What an opening

[LordSnooty]

Because of increasing awareness in the general public about energy conservation, the ability to utilize low power states on desktop PCs is incredibly underdocumented and widely unused.

The opening sentence fails to compile in my logic parser - there is little documentation because of increasing awareness? Better would have been: "Because of increasing awareness in the general public about energy conservation, people want to know more about the ability to utilize low power states on desktop PCs. What they're finding is that it's incredibly underdocumented and widely unused." Oh, and "underdocumented" doesn't appear to be a word.

Welcome to the exciting new world of UGC.

In before the morons

[imsabbel]

that will now cry that their computer _has_ to be on, 24/7 (because otherwise, they couldnt improve their epenis, er, i mean uptime).

Well, i used to have my computer always on. When it wasnt needed, it went into hybernation mode. After i upgraded my system (changed from amd and old socket 939 to core2. Btw, against the "wisdom" of the typical moron, without reinstalling windows. Isnt needed since win2000, people...), S3 was activated automatically.

And it NEVER made any problems. The limiting factor for wake

A nice way to rub salt in a wound.

[twitter]

that will now cry that their computer _has_ to be on, 24/7 (because otherwise, they couldnt improve their epenis, er, i mean uptime).

Uptime is something M$ can't deliver, but they have done a nice job of making sure systems with good uptime can't do power management easily [edge-op.org]. ACPI is sabotaged. I'd love to be able to have all my systems hibernate AND be network accessable, but I have not had the time to see that it does or does not work on my system yet.

From what I've heard, M$ has also unable to delive

re: S3 Standby State Done Right

[AncientPC]

Slightly off tangent, but hibernation (S4) fails in WinXP SP2 if you have more than 1GB of RAM. [microsoft.com]

My biggest problem with standby on my WinXP machine is that my machine will randomly wake up after a random amount of time. I've already disabled WOL and Wake-on-USB, but my computer will wake up randomly from standby anywhere from 3 minutes to never. I still can't figure out what's causing the problem. :(

Re: S3 Standby State Done Right

[bi_boy]

Slightly off tangent, but hibernation (S4) fails in WinXP SP2 if you have more than 1GB of RAM. [microsoft.com]

Works just fine for me. Probably because I installed the udpate mentioned in the resolution section of the article sometime last year.

Re:

[jesdynf]

Video drivers. I've seen it before -- switch to an ATI card and you'll be fine.

Those nSomnia cards cause nothing but grief.

I'd seriously consider blaming the mouse. Try unplugging it before sending the computer into standby, and see what happens. Micromovements are always what cause /my/ machines to somnanbulate.

Re:

[DeafByBeheading]

Don't bet on it. I have an ATI card (X1600) and I had that problem. Patch worked for me.

Re:

[jesdynf]

Err. Having read your serious reply to my worthless suggestion, I see that Slashdot helpfully deleted the (non-conforming) markup I used to indicate what was a joke and what wasn't. Phooey. Shoulda caught that in preview, I suppose...

Programmatically disabling sleep mode on Vista.

[Vellmont]

I usually have sleep mode enabled after 60 minutes of inactivity, but I occasionally need to keep the machine up because I'm downloading something.

Ideally I'd like to recompile the app with code inserted to disable sleep mode while it's running, or register some kind of user activity. Anyone know how this is accomplished on Vista? Another option would be to write another app that monitors network activity, and disables sleep mode when it sees activity over a certain threshold.

Re:

[PhrostyMcByte]

You can call SetThreadExecutionState() periodically to disable automatic sleeping. It does not prevent manual sleeping though.

ACPI is a clusterfuck

[bersl2]

Everybody who believes to the contrary should have to dig around in disassembled DSDT code to try to make their thermal zones report the temperature properly (which I've done) and make the machine resume from S3 instead of rebooting when hitting the power button (which has failed miserably) in a spec-compliant ACPI implementation.

If you haven't done that, go do it. Now. Gain an appreciation for either the difficulty in following the spec properly or the incompetence of the OEMs to implement their part prope

sounds like a great way to wreck disks

[Loconut1389]

S3 standby shuts off hard drives and if you're running a file-server, aside from the lag people will experience between their packet and your computer reviving itself, power cycling your drives dozens of times a day will greatly reduce their lifespan. If you're worried about going green, buying an unnecessary hard drive probably puts more chemicals into the environment from its manufacturing than leaving your computer on?

To see available Sleep modes from Windows Prompt

[icepick72]

From command window:

powercfg -a
Works for both XP and Vista. Tells you what's available and what's not (S1, S2, S3,...) Vista tells you why something isn't support.

Got info from this page [tech-recipes.com]

In my expirence power saving in Windows crashes

[Orion Blastar]

the system. I followed directions like that S3 article, and what I get is a Windows that locks up and won't wake up. I don't know if it is a hardware, OS, or software issue. I usually get the login screen and then the mouse cursor won't move and the keyboard no longer works and beeps at me for each key I press.

My laptop works that way and so does my desktop. It is very annoying and in order to avoid the lockups I have to turn power saving off and then the system doesn't lock up anymore.

Re:

[gEvil (beta)]

I haven't used any of the MacPro's, but in the PPC era the sleep state was similar to S3 standby.

Re:

[ettlz]

Back in the pre-NT-based days, perhaps. Most modern operating system kernels issue the HLT instruction plus some extra power management jiggery-pokery to the CPU when it's not being used at max anyway. (Check out /proc/acpi/processor/CPU*/power under Linux.)

Re:

[Blahbooboo3]

I know you are trying to http://en.wikipedia.org/wiki/Internet_troll [wikipedia.org] troll, but is that the best you can do? This has got to be one of the lamest troll comments in a while on slashdot...

Put some effort into it at least! Let me get you started with a helpful guide to trolling http://www.urban75.com/Mag/troll.html [urban75.com]