I'm planning on getting a home server, serving httpd, ssh, mysql, dovecot, etc, and running Fedora 7.
I'd like something using as little energy as possible, given that I need 200-300GB in disk-space. But I don't need a monitor (or graphics card).
Is anyone running such a system? If so, I'd be interested to hear the hardware details.
On Sat, 29 Sep 2007, Timothy Murphy wrote:
I'm planning on getting a home server, serving httpd, ssh, mysql, dovecot, etc, and running Fedora 7.
I'd like something using as little energy as possible, given that I need 200-300GB in disk-space. But I don't need a monitor (or graphics card).
Is anyone running such a system? If so, I'd be interested to hear the hardware details.
I have been running for four years a VIA C3 800Mhz computer, with 512MB of RAM, and a pair of 80GB IDE drives. It pulls 45 watts from the wall when idle - and 55 when 'crunching' away.
A current system could be a PC-Chips M789CG motherboard.
Sure its not fast, but works well as a university department listserve for the students, a web site running moodle, and a file server for the staff.
For me, it was the lack of noise from the box that drove me to the VIA C3. There is a passive heat sync on the CPU - with a quite case fan clearing the heat.
I suspect that once the drives spin up, they don't draw any more then 9-10 watts.
Also, match the power supply to the load. If you only need 50 watts, do not buy a 400 watt power supply. I think I have a 150 or 200 watt power supply in my VIA C3 computer, LPX form factor.
ed
ed@hp.uab.edu wrote:
I have been running for four years a VIA C3 800Mhz computer, with 512MB of RAM, and a pair of 80GB IDE drives. It pulls 45 watts from the wall when idle - and 55 when 'crunching' away.
Thanks for the suggestions.
I've been a little surprised, looking at a few setups, how little energy one can save. My incredibly ancient PIII with 2 SCSI disks and one IDE disk, seems to use just over 80 watts in normal usage. I was sort of expecting I'd be able to get this down to 20-30 watts, but the really low power systems I've looked at all seem deficient in one way or another.
On Sun, 30 Sep 2007, Timothy Murphy wrote:
ed@hp.uab.edu wrote:
I have been running for four years a VIA C3 800Mhz computer, with 512MB of RAM, and a pair of 80GB IDE drives. It pulls 45 watts from the wall when idle - and 55 when 'crunching' away.
Thanks for the suggestions.
I've been a little surprised, looking at a few setups, how little energy one can save. My incredibly ancient PIII with 2 SCSI disks and one IDE disk, seems to use just over 80 watts in normal usage. I was sort of expecting I'd be able to get this down to 20-30 watts, but the really low power systems I've looked at all seem deficient in one way or another.
Well don't give up so fast. For example, Western Digital offers the Carvier GP line of drives that idle at 4 watts. And you could to use 2.5 inch drives from laptops, the Scorpio from Western Digital idles at 2 watts.
I would consider spinning down the drives, only if I had laptop drives. They are small and light - so much less wear and tear.
Another option is Flash, for the '/' partition. Then laptop drives for the data. 4GB Compact flash media is about $50. You could make the 20 watts.
ed
On Sat, 2007-09-29 at 18:43 -0400, ed@hp.uab.edu wrote:
Also, match the power supply to the load. If you only need 50 watts, do not buy a 400 watt power supply. I think I have a 150 or 200 watt power supply in my VIA C3 computer, LPX form factor.
While you don't need a huge supply, one bigger than your current needs gives you more room to play with later on, if you have to add something, or the power supply gets a bit sick. The wattage is the capacity of what it can deliver, not that it's always going to draw 400 Watts from the mains and be a drain on your electricity bill.
On Mon, 1 Oct 2007, Tim wrote:
On Sat, 2007-09-29 at 18:43 -0400, ed@hp.uab.edu wrote:
Also, match the power supply to the load. If you only need 50 watts, do not buy a 400 watt power supply. I think I have a 150 or 200 watt power supply in my VIA C3 computer, LPX form factor.
While you don't need a huge supply, one bigger than your current needs gives you more room to play with later on, if you have to add something, or the power supply gets a bit sick. The wattage is the capacity of what it can deliver, not that it's always going to draw 400 Watts from the mains and be a drain on your electricity bill.
You are correct, that at 400 watt power supply does not always draw 400 watts from the wall. But the efficiancy of the power supply is less when how running at full power. Take this article for example:
http://www.silentpcreview.com/article263-page4.html
The 300 watt power supply leaks 12 watts when the computer only needs 40 watts.
I may be working on a project that needs to run in a car, but not an entertainment system. Some links of interest for this thread:
motherboard / cpu: JetWay J627F800-OC VIA C3 processor VIA CLE266 Mini ITX Motherboard/CPU Combo
cases and power supplies: http://www.mini-box.com You can find a 60 watt power supply (12V 5A) to run a system
ed
ed@hp.uab.edu wrote:
Also, match the power supply to the load. If you only need 50 watts, do not buy a 400 watt power supply. I think I have a 150 or 200 watt power supply in my VIA C3 computer, LPX form factor.
While you don't need a huge supply, one bigger than your current needs gives you more room to play with later on, if you have to add something, or the power supply gets a bit sick. The wattage is the capacity of what it can deliver, not that it's always going to draw 400 Watts from the mains and be a drain on your electricity bill.
You are correct, that at 400 watt power supply does not always draw 400 watts from the wall. But the efficiancy of the power supply is less when how running at full power. Take this article for example:
http://www.silentpcreview.com/article263-page4.html
The 300 watt power supply leaks 12 watts when the computer only needs 40 watts.
Probably an ignorant question, but if one is talking about a mini-ITX system, will a cheap external AC power adaptor do the job, like those incorporated into a power plug, eg http://www.mini-box.com/60w-12v-5A-AC-DC-Power-Adapter_2?sc=8&category=13?
On Mon, 1 Oct 2007, Timothy Murphy wrote:
ed@hp.uab.edu wrote:
Also, match the power supply to the load. If you only need 50 watts, do not buy a 400 watt power supply. I think I have a 150 or 200 watt power supply in my VIA C3 computer, LPX form factor.
While you don't need a huge supply, one bigger than your current needs gives you more room to play with later on, if you have to add something, or the power supply gets a bit sick. The wattage is the capacity of what it can deliver, not that it's always going to draw 400 Watts from the mains and be a drain on your electricity bill.
You are correct, that at 400 watt power supply does not always draw 400 watts from the wall. But the efficiancy of the power supply is less when how running at full power. Take this article for example:
http://www.silentpcreview.com/article263-page4.html
The 300 watt power supply leaks 12 watts when the computer only needs 40 watts.
Probably an ignorant question, but if one is talking about a mini-ITX system, will a cheap external AC power adaptor do the job, like those incorporated into a power plug, eg http://www.mini-box.com/60w-12v-5A-AC-DC-Power-Adapter_2?sc=8&category=13?
Well not quite. Its OK if you don't have any spinning platters in your hard drive. But you asked for 200-300GB of storage.
But if you just wanted 32GB of flash... then this would work.
Or, you had all your big drives on the usb 2.0 bus.
ed
ed@hp.uab.edu wrote:
Probably an ignorant question, but if one is talking about a mini-ITX system, will a cheap external AC power adaptor do the job, like those incorporated into a power plug, eg
http://www.mini-box.com/60w-12v-5A-AC-DC-Power-Adapter_2?sc=8&category=13?
Well not quite. Its OK if you don't have any spinning platters in your hard drive. But you asked for 200-300GB of storage.
But if you just wanted 32GB of flash... then this would work.
Or, you had all your big drives on the usb 2.0 bus.
Well, this is a 60 watt adaptor, not 30 watts as someone said. I would have thought that was quite enough for a Mini-ITX + one disk. My ancient PIII system, with 3 disks (including 300MB SCSI and 250MB IDE drives) and no care taken about power consumption, only takes 80-90 watts, even at startup. This system has 2 fans (CPU and box) and a 250 watt PSU.
I'm interested in a system running off 240 volts, not an in-car system. Wouldn't something like the fanless VIA EPIA EN12000E with a low-power laptop disk (maybe 100-150GB) do the job?
I'm surprised how little discussion of what I would have thought was a common need - a low-power server - there seems to be.
On Mon, 2007-10-01 at 21:45 +0100, Timothy Murphy wrote:
ed@hp.uab.edu wrote:
Probably an ignorant question, but if one is talking about a mini-ITX system, will a cheap external AC power adaptor do the job, like those incorporated into a power plug, eg
http://www.mini-box.com/60w-12v-5A-AC-DC-Power-Adapter_2?sc=8&category=13?
Well not quite. Its OK if you don't have any spinning platters in your hard drive. But you asked for 200-300GB of storage.
But if you just wanted 32GB of flash... then this would work.
Or, you had all your big drives on the usb 2.0 bus.
Well, this is a 60 watt adaptor, not 30 watts as someone said. I would have thought that was quite enough for a Mini-ITX + one disk. My ancient PIII system, with 3 disks (including 300MB SCSI and 250MB IDE drives) and no care taken about power consumption, only takes 80-90 watts, even at startup. This system has 2 fans (CPU and box) and a 250 watt PSU.
I'm interested in a system running off 240 volts, not an in-car system. Wouldn't something like the fanless VIA EPIA EN12000E with a low-power laptop disk (maybe 100-150GB) do the job?
I'm surprised how little discussion of what I would have thought was a common need - a low-power server - there seems to be.
---- I think that you are more or less unique in describing low power consumption as a priority for a server.
Server grade components tend to be constructed for more rugged continuous usage. While you can obviously succeed in obtaining low power hardware, you will lose much in terms of redundancy, over engineering for wider tolerances, higher internal temperatures, etc.
Craig White wrote:
I'm surprised how little discussion of what I would have thought was a common need - a low-power server - there seems to be.
I think that you are more or less unique in describing low power consumption as a priority for a server.
I wonder if that is true? Maybe the word "server" is used in two different senses. I'm thinking of a "home server" - one that links to the internet, and serves a few laptops on a small wifi or ethernet lan. This is on all the time, so power usage will mount up.
I think the word server was used historically for a computer serving a large number of terminals or nowadays laptops, perhaps in a small business or university department. In this case, as you say, power usage will come way down in the order of priorities, after reliability and related issues.
Server grade components tend to be constructed for more rugged continuous usage. While you can obviously succeed in obtaining low power hardware, you will lose much in terms of redundancy, over engineering for wider tolerances, higher internal temperatures, etc.
As I said, we are thinking of different scenarios. I've been quite struck by the fact that my ancient PIII home server seems quite adequate for the task - running httpd, ssh, dovecot, collecting email, etc.
It seems to me that a mini-ITX with one SATA laptop disk should be able to do the work without difficulty. I'm not sure if a fanless VIA EPIA system, or something similar, could contain a disk, though?
Regarding RAID, I wonder if that is worth it for such a system, if one does nightly backups? I see that disks now are often guaranteed for 5 years, which must mean the half-life is something more than this. I would rank other possible disasters - fire, flood, etc, far higher than this - always as I said in a home setting. And it wouldn't be that difficult to change the disk if it failed.
Timothy Murphy wrote:
ed@hp.uab.edu wrote:
Also, match the power supply to the load. If you only need 50 watts, do not buy a 400 watt power supply. I think I have a 150 or 200 watt power supply in my VIA C3 computer, LPX form factor.
While you don't need a huge supply, one bigger than your current needs gives you more room to play with later on, if you have to add something, or the power supply gets a bit sick. The wattage is the capacity of what it can deliver, not that it's always going to draw 400 Watts from the mains and be a drain on your electricity bill.
You are correct, that at 400 watt power supply does not always draw 400 watts from the wall. But the efficiancy of the power supply is less when how running at full power. Take this article for example:
http://www.silentpcreview.com/article263-page4.html
The 300 watt power supply leaks 12 watts when the computer only needs 40 watts.
Probably an ignorant question, but if one is talking about a mini-ITX system, will a cheap external AC power adaptor do the job, like those incorporated into a power plug, eg http://www.mini-box.com/60w-12v-5A-AC-DC-Power-Adapter_2?sc=8&category=13?
I'm an EE and I know that every power supply made today for computers is one of the switching types. They ALL have a device designed to switch any AC voltage from about 90 to 240 VAC.
The switched waveform is very fast. We use 60 Hz here and this power supply switches at 16,000 Hz. Some nice things happen at that high frequency, like transformers get small. So you can make a transformer that puts out the proper voltage to make plus 5 volts and plus 12 volts and others that are regulated.
If you measure the power input to a switcher with no load you will discover it's quite low, like 30-40 watts. This is wasted power just to run the power supply with no output. But if you put a computer load on the supply it will be directly sent to the AC load but with very small loss.
So if you buy a 450 Watt power supply to run a server that does nothing most of the time, the power supply will not draw 450 watts. It will draw perhaps 100 Watts. This is good. The power supply is lightly loaded and will last much longer than if it is drawing 400 watts.
So do not worry about what the power supply is rated at. It will adjust to a smaller load.
On Mon, 1 Oct 2007, Karl Larsen wrote:
So if you buy a 450 Watt power supply to run a server that does nothing most of the time, the power supply will not draw 450 watts. It will draw perhaps 100 Watts. This is good. The power supply is lightly loaded and will last much longer than if it is drawing 400 watts.
So do not worry about what the power supply is rated at. It will adjust to a smaller load.
Everything you said was true, and it still does not help to have a lightly loaded 450 watt power supply drawing 100 Watts... If your target is to draw 30 watts from the wall. In fact you validated my point that matching the power supply is better then over buying excess capacity.
ed
p.s. You're not the only one here with an EE degree.
On Mon, 2007-10-01 at 11:11 -0400, ed@hp.uab.edu wrote:
On Mon, 1 Oct 2007, Karl Larsen wrote:
So if you buy a 450 Watt power supply to run a server that does nothing most of the time, the power supply will not draw 450 watts. It will draw perhaps 100 Watts. This is good. The power supply is lightly loaded and will last much longer than if it is drawing 400 watts.
So do not worry about what the power supply is rated at. It will adjust to a smaller load.
Everything you said was true, and it still does not help to have a lightly loaded 450 watt power supply drawing 100 Watts... If your target is to draw 30 watts from the wall. In fact you validated my point that matching the power supply is better then over buying excess capacity.
ed
p.s. You're not the only one here with an EE degree.
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current.
Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Les wrote:
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current.
Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Here's some more interesting information on power supplies. Note that best efficiency is attained at/near maximum load. So a 500 watt supply looks good but may not be what you want when considering power consumption and disposing of waste heat.
http://services.google.com/blog_resources/PSU_white_paper.pdf
Bob Goodwin
Bob Goodwin wrote:
Les wrote:
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current. Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Here's some more interesting information on power supplies. Note that best efficiency is attained at/near maximum load. So a 500 watt supply looks good but may not be what you want when considering power consumption and disposing of waste heat.
http://services.google.com/blog_resources/PSU_white_paper.pdf
Bob Goodwin
Hi Bob, I am interested in what Google has done but need a lot more information. They say changing the PS to one making just plus 12 volts will save energy. I do not see that as a given. In fact it makes me think they are just passing on the voltage change to the mother-board makers. I think the idea being the MB makers can make just the other voltages their board needs. But converting 12 to 1.5 volts with economy of power is complex and will jack up the MB cost somewhat, and will make it hotter.
On 10/1/07, Karl Larsen k5di@zianet.com wrote:
Bob Goodwin wrote:
Les wrote:
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current. Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Here's some more interesting information on power supplies. Note that best efficiency is attained at/near maximum load. So a 500 watt supply looks good but may not be what you want when considering power consumption and disposing of waste heat.
http://services.google.com/blog_resources/PSU_white_paper.pdf
Bob Goodwin
Hi Bob, I am interested in what Google has done but need a lot moreinformation. They say changing the PS to one making just plus 12 volts will save energy. I do not see that as a given. In fact it makes me think they are just passing on the voltage change to the mother-board makers. I think the idea being the MB makers can make just the other voltages their board needs. But converting 12 to 1.5 volts with economy of power is complex and will jack up the MB cost somewhat, and will make it hotter.
--
Karl F. Larsen, AKA K5DI Linux User #450462 http://counter.li.org.
They're trying to get rid of distributing AC to all the racks and the losses resulting converting from AC to DC for each MB. If the MB design only requires 12 VDC input then it is more efficient to have only one AC to DC converter and buss the DC to all the boxes in a rack. By the way, people have been bussing 24 and 48 VDC in equipment racks for years.
Kam Leo wrote:
On 10/1/07, Karl Larsen k5di@zianet.com wrote:
Bob Goodwin wrote:
Les wrote:
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current. Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Here's some more interesting information on power supplies. Note that best efficiency is attained at/near maximum load. So a 500 watt supply looks good but may not be what you want when considering power consumption and disposing of waste heat.
http://services.google.com/blog_resources/PSU_white_paper.pdf
Bob Goodwin
Hi Bob, I am interested in what Google has done but need a lot moreinformation. They say changing the PS to one making just plus 12 volts will save energy. I do not see that as a given. In fact it makes me think they are just passing on the voltage change to the mother-board makers. I think the idea being the MB makers can make just the other voltages their board needs. But converting 12 to 1.5 volts with economy of power is complex and will jack up the MB cost somewhat, and will make it hotter.
--
Karl F. Larsen, AKA K5DI Linux User #450462 http://counter.li.org.They're trying to get rid of distributing AC to all the racks and the losses resulting converting from AC to DC for each MB. If the MB design only requires 12 VDC input then it is more efficient to have only one AC to DC converter and buss the DC to all the boxes in a rack. By the way, people have been bussing 24 and 48 VDC in equipment racks for years.
The 12 volts can not be moved as 12 volts DC very far. The length of a car is about it. So Google must have a lot of racks in close proximity.
On Mon, 2007-10-01 at 12:54 -0600, Karl Larsen wrote:
Kam Leo wrote:
On 10/1/07, Karl Larsen k5di@zianet.com wrote:
Bob Goodwin wrote:
Les wrote:
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current. Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Here's some more interesting information on power supplies. Note that best efficiency is attained at/near maximum load. So a 500 watt supply looks good but may not be what you want when considering power consumption and disposing of waste heat.
http://services.google.com/blog_resources/PSU_white_paper.pdf
Bob Goodwin
Hi Bob, I am interested in what Google has done but need a lot moreinformation. They say changing the PS to one making just plus 12 volts will save energy. I do not see that as a given. In fact it makes me think they are just passing on the voltage change to the mother-board makers. I think the idea being the MB makers can make just the other voltages their board needs. But converting 12 to 1.5 volts with economy of power is complex and will jack up the MB cost somewhat, and will make it hotter.
Snip sig.
They're trying to get rid of distributing AC to all the racks and the losses resulting converting from AC to DC for each MB. If the MB design only requires 12 VDC input then it is more efficient to have only one AC to DC converter and buss the DC to all the boxes in a rack. By the way, people have been bussing 24 and 48 VDC in equipment racks for years.
The 12 volts can not be moved as 12 volts DC very far. The length ofa car is about it. So Google must have a lot of racks in close proximity.
And while moving this single voltage around has lots of advantages for racks, it is not such a clear advantage for single systems. I guess only time will tell us that story.
What Karl has seen in working on Amatuer Radio, and my experience on a boat tells us the story on high current busing which is a very complex, very hazardous proposition. And yes I know that it is done all the time in racks, and along buses, however, check out the AWG specifications for 12v with 10% drop at 30' and think about the amount of copper needed for a data center. Up to about 5 racks is probably acceptable, but beyond that the copper cost and the copper losses begin to mount quite rapidly, especially for hard drives or other motor driven equipment. Anything that generates force, motors, relays, driver circuits for lasers, lamps etc. force the current requirements quite rapidly. In addition, as the voltage drops, the area of the conductor goes up by the square. This establishes a point of optimum economic return, and also establishes the acceptable connector loss and of course the resultant heat and fire potential, which are primarily current dependent (a 500Watt soldering gun actually uses only about 1.2V at its tip loop for example.)
Then there is another whole subject that I touched on, that of power factor. Generally AC loads develop maximum power at 90degrees between current and voltage. But some components, such as totally passive ones, develop maximum power and efficiency at 0 degrees difference between current and voltage. The variance between 0 and 90 degrees determines the efficiency of power transfer. So a very efficient device at 0 degrees may only be 30% efficient at 90 degrees and vice-versa. However if the lines are loaded with a 0 degree phase, the copper losses mount and the wires carrying AC begin to heat up. This increases losses accelerating heating... a condition known as thermal runaway, and can result in catastrophic failure of a power delivery conduit. Not good. So as the world matures, and develops more and more electrical loads, we are all having to learn more and more about power and AC functionality in the engineering community. Lessons the RF guys have known for years are now beginning to affect the digital and power folks and the colleges are struggling to bring all this into the curriculum in a manner that reflects good design practices as well.
I spent much of my life working on RF, so I find it interesting, but not too demanding to apply these principals to AC power. Those without a background in RF power electronics occasionally find themselves struggling to get a handle on the implications.
Regards, Les H
On Monday 01 October 2007, Les wrote:
On Mon, 2007-10-01 at 12:54 -0600, Karl Larsen wrote:
The 12 volts can not be moved as 12 volts DC very far. The length ofa car is about it. So Google must have a lot of racks in close proximity.
And while moving this single voltage around has lots of advantages for racks, it is not such a clear advantage for single systems. I guess only time will tell us that story.
What Karl has seen in working on Amatuer Radio, and my experience on a boat tells us the story on high current busing which is a very complex, very hazardous proposition.
I have spent 17 years as a broadcast engineer (the last four of which just consulting); RF is one of my things.
So is DC power.
The really nice thing about DC power is the complete lack of skin effect. That is, once you are talking lots and lots of current, and needing the lots and lots of copper to carry it, DC will use 100% of your copper, where even 60Hz AC will not. The ampacity for, say, 750kcmil copper wire, RHHW or XHHW preferably, at DC is much much higher than at AC.
12VDC is too low of a voltage to take advantage of this; 48VDC, on the other hand, is just about the sweet spot, as long as you are in the 200A or greater club. We are running 48VDC distribution for switches and servers here, and it's pretty interesting that a server that will take a 15A circuit at AC due to power factor correction things and inrush currents will take a smaller DC circuit, and the same wire size at 48VDC. The larger switches and routers that I have (Cisco 12012, 8540, and 7507) are a little different; but I can use not that much larger wire to run them (10 or 8AWG x2 instead of 12 or 10 x3, typically).
The conversion efficiency is not the driver for me; the driver is the centralized UPS and single large (450Ah@ 48VDC nominal) battery that runs everything. UPS inverter losses are not insignificant in a data center; 12VDC is one of the three really useful voltages to use (24 and 48 being the other two) where battery backup really helps.
AC distribution to modern power supplies, where a UPS isn't an issue, isn't any less efficient, as the typical switching power supply's input circuit is a simple rectifier to get 160VDC, which is then DC-DC converted to the various voltages.
Les wrote:
On Mon, 2007-10-01 at 12:54 -0600, Karl Larsen wrote:
Kam Leo wrote:
On 10/1/07, Karl Larsen k5di@zianet.com wrote:
Bob Goodwin wrote:
Les wrote:
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current. Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Here's some more interesting information on power supplies. Note that best efficiency is attained at/near maximum load. So a 500 watt supply looks good but may not be what you want when considering power consumption and disposing of waste heat.
http://services.google.com/blog_resources/PSU_white_paper.pdf
Bob Goodwin
Hi Bob, I am interested in what Google has done but need a lot moreinformation. They say changing the PS to one making just plus 12 volts will save energy. I do not see that as a given. In fact it makes me think they are just passing on the voltage change to the mother-board makers. I think the idea being the MB makers can make just the other voltages their board needs. But converting 12 to 1.5 volts with economy of power is complex and will jack up the MB cost somewhat, and will make it hotter.
Snip sig.
They're trying to get rid of distributing AC to all the racks and the losses resulting converting from AC to DC for each MB. If the MB design only requires 12 VDC input then it is more efficient to have only one AC to DC converter and buss the DC to all the boxes in a rack. By the way, people have been bussing 24 and 48 VDC in equipment racks for years.
The 12 volts can not be moved as 12 volts DC very far. The length ofa car is about it. So Google must have a lot of racks in close proximity.
And while moving this single voltage around has lots of advantages for racks, it is not such a clear advantage for single systems. I guess only time will tell us that story.
What Karl has seen in working on Amatuer Radio, and my experience on a boat tells us the story on high current busing which is a very complex, very hazardous proposition. And yes I know that it is done all the time in racks, and along buses, however, check out the AWG specifications for 12v with 10% drop at 30' and think about the amount of copper needed for a data center. Up to about 5 racks is probably acceptable, but beyond that the copper cost and the copper losses begin to mount quite rapidly, especially for hard drives or other motor driven equipment. Anything that generates force, motors, relays, driver circuits for lasers, lamps etc. force the current requirements quite rapidly. In addition, as the voltage drops, the area of the conductor goes up by the square. This establishes a point of optimum economic return, and also establishes the acceptable connector loss and of course the resultant heat and fire potential, which are primarily current dependent (a 500Watt soldering gun actually uses only about 1.2V at its tip loop for example.)
Then there is another whole subject that I touched on, that of power factor. Generally AC loads develop maximum power at 90degrees between current and voltage. But some components, such as totally passive ones, develop maximum power and efficiency at 0 degrees difference between current and voltage. The variance between 0 and 90 degrees determines the efficiency of power transfer. So a very efficient device at 0 degrees may only be 30% efficient at 90 degrees and vice-versa. However if the lines are loaded with a 0 degree phase, the copper losses mount and the wires carrying AC begin to heat up. This increases losses accelerating heating... a condition known as thermal runaway, and can result in catastrophic failure of a power delivery conduit. Not good. So as the world matures, and develops more and more electrical loads, we are all having to learn more and more about power and AC functionality in the engineering community. Lessons the RF guys have known for years are now beginning to affect the digital and power folks and the colleges are struggling to bring all this into the curriculum in a manner that reflects good design practices as well.
I spent much of my life working on RF, so I find it interesting, but not too demanding to apply these principals to AC power. Those without a background in RF power electronics occasionally find themselves struggling to get a handle on the implications.
Regards, Les H
Hi Les the typical measure of by how much the current leads or lags the voltage is called "power factor". This is VERY important to power providers and they spend big bucks for Capacitors to put across the line to try and keep power factor = 1. If the power factor gets to around 0.5 and the power company is wasting a lot of power providing current that is wasted.
50 years ago I worked for Hughes and was put in charge of the MA1 fire control power supply for the F-106 airplane. I was told the 1600 Hz power source was running hot. I went to the factory to the system running and found the right cable and measured the power factor and found it was .06 about, inductive. I got a worker to put in a .8 UFD capacitor which when I measured pf again it was .95 and the system cooled down.
So for AC power performance you want the current and voltage to be in phase.
On 10/1/07, Karl Larsen k5di@zianet.com wrote:
Kam Leo wrote:
On 10/1/07, Karl Larsen k5di@zianet.com wrote:
Bob Goodwin wrote:
Les wrote:
These supplies are switching supplies, not linear. You should check some of the draws, I think you will be surprised about how little some of them draw (admittedly there are various design strengths out there.)
In general, a linear supply is about 20-35% efficient, meaning that a 450 watt supply would draw about 90 watts all the time. But these new switchers are about double that or more, meaning that the supply will draw somewhere between 20 and 50 watts. Moreover that draw will basically be independent of the size of the ultimate supply, simply existing to control the switcher itself, and not the drive current. Technology exists in some forms of "buck control" that could boost the on line efficiency further reducing the drain. Power stuff is evolving even faster than much of the other technology, but is less glamorous, so doesn't get as much of the press.
Regards, Les H
Here's some more interesting information on power supplies. Note that best efficiency is attained at/near maximum load. So a 500 watt supply looks good but may not be what you want when considering power consumption and disposing of waste heat.
http://services.google.com/blog_resources/PSU_white_paper.pdf
Bob Goodwin
Hi Bob, I am interested in what Google has done but need a lot moreinformation. They say changing the PS to one making just plus 12 volts will save energy. I do not see that as a given. In fact it makes me think they are just passing on the voltage change to the mother-board makers. I think the idea being the MB makers can make just the other voltages their board needs. But converting 12 to 1.5 volts with economy of power is complex and will jack up the MB cost somewhat, and will make it hotter.
--
Karl F. Larsen, AKA K5DI Linux User #450462 http://counter.li.org.They're trying to get rid of distributing AC to all the racks and the losses resulting converting from AC to DC for each MB. If the MB design only requires 12 VDC input then it is more efficient to have only one AC to DC converter and buss the DC to all the boxes in a rack. By the way, people have been bussing 24 and 48 VDC in equipment racks for years.
The 12 volts can not be moved as 12 volts DC very far. The length ofa car is about it. So Google must have a lot of racks in close proximity.
It's probably more like one AC/DC converter per rack. Similar to a blade system where there is one power supply sending regulated power to blades which plug into the main chassis.
On Monday 01 October 2007, Karl Larsen wrote:
But converting 12 to 1.5 volts with economy of power is complex and will jack up the MB cost somewhat, and will make it hotter.
Actually, this is already being done in your PC now. PC power supplies only generate 3.3V, 5V, and 12V; they don't directly generate the CPU and RAM voltages.
Beginning with the Pentium 4 and Athlon 64 chips, most motherboards use a 12V input to the CPU core regulators; prior motherboards either used the 3.3V or 5V outputs of the power supply regulated down to the CPU core voltage.
There are little 12VDC 'paddleboards' now that plug into an ATX motherboard plug; 200W units are very small. See the units at mini-box.com for comparisons; the PSU board is part of the ATX plug itself in most cases.
ed@hp.uab.edu wrote:
On Mon, 1 Oct 2007, Karl Larsen wrote:
So if you buy a 450 Watt power supply to run a server that does nothing most of the time, the power supply will not draw 450 watts. It will draw perhaps 100 Watts. This is good. The power supply is lightly loaded and will last much longer than if it is drawing 400 watts.
So do not worry about what the power supply is rated at. It will adjust to a smaller load.
Everything you said was true, and it still does not help to have a lightly loaded 450 watt power supply drawing 100 Watts... If your target is to draw 30 watts from the wall. In fact you validated my point that matching the power supply is better then over buying excess capacity.
ed
p.s. You're not the only one here with an EE degree.
I know that. I was not aware the goal was to draw 30 Watts or less. I do not think you can do that with a power supply mass produced for computers. You need more than that to boot up the system. The hard drives use like 20 watts when loaded. If you have two working it will be too much for a power source you are trying to use. Forget that 30 Watt limit. It will not work.
On Mon, 1 Oct 2007, Karl Larsen wrote:
I know that. I was not aware the goal was to draw 30 Watts or less. I do not think you can do that with a power supply mass produced for computers. You need more than that to boot up the system. The hard drives use like 20 watts when loaded. If you have two working it will be too much for a power source you are trying to use. Forget that 30 Watt limit. It will not work.
Well, we don't just give up because it sounds hard. Timothy, the original poster, needs to decide if its worth the money and time to take a server down to 30 watts.
I'm interested, because it looks like there is an ability to run a computer with 5 watts. Its just a matter of what I will be able to do with it.
http://www.mini-box.com/Alix-1C-Board-1-LAN-1-MINI-PCI
ed
ed@hp.uab.edu wrote:
On Mon, 1 Oct 2007, Karl Larsen wrote:
I know that. I was not aware the goal was to draw 30 Watts or less. I do not think you can do that with a power supply mass produced for computers. You need more than that to boot up the system. The hard drives use like 20 watts when loaded. If you have two working it will be too much for a power source you are trying to use. Forget that 30 Watt limit. It will not work.
Well, we don't just give up because it sounds hard. Timothy, the original poster, needs to decide if its worth the money and time to take a server down to 30 watts.
I'm interested, because it looks like there is an ability to run a computer with 5 watts. Its just a matter of what I will be able to do with it.
http://www.mini-box.com/Alix-1C-Board-1-LAN-1-MINI-PCI
ed
Hi Ed, right now the CPU cooler fan on this computer draws more than 5 Watts :-)
I have the manual for the 5 watt computer and will read it. I see they power the computer from a car battery. So the whole thing is taking just w = v times i so i = w/v = 25/12 = 2 amps. That is a tiny draw for a new car. My old Durango has a 125 Amp alternator!
On Mon, 2007-10-01 at 09:21 -0400, ed@hp.uab.edu wrote:
at 400 watt power supply does not always draw 400 watts from the wall. But the efficiancy of the power supply is less when how running at full power.
That really depends on the design. Nothing's 100% efficient, some are better than others. Whilst true that they probably optimise a supply for a certain range, I doubt that there's a really great difference between most of the switch-mode power supplies, they're all mostly the same circuitry, with bigger caps and beefier transistors for the higher powered ones (so they'll last a higher duty cycle), and you'd still want to pick one that offers more than you're going to need.
You'd also want to look carefully at the specs, you'll probably find many power supplies rated at what they can supply, but not mentioning how much they draw. I suspect that you'd have to look for special high efficiency units, rather than simply looking for lower wattage ones, as they're probably all rather poor.
My beef with many power supplies is that they draw too much all of the time. We can't easily turn off computers, just shut them down. It's a pain to go crawling under desks and around racks to reach the power points. And you don't want someone turning off the wrong point.
On Sat, 2007-09-29 at 22:42 +0100, Timothy Murphy wrote:
I'm planning on getting a home server, serving httpd, ssh, mysql, dovecot, etc, and running Fedora 7.
I'd like something using as little energy as possible, given that I need 200-300GB in disk-space. But I don't need a monitor (or graphics card).
Is anyone running such a system? If so, I'd be interested to hear the hardware details.
---- I have 3 shuttle SFF (small form factor) systems and a Macintosh on a single 4 port KVM switch.
The Shuttle SFF only allows 1 hard drive and 1 CD (or I suppose you could use a second hard drive instead of CD) but drives are available up to 750 Gb now (perhaps higher but I haven't seen one).
If I go to www.shuttle.com, it automatically takes me to their US site - YMMV
Craig
On Sat, 2007-09-29 at 22:42 +0100, Timothy Murphy wrote:
I'm planning on getting a home server, serving httpd, ssh, mysql, dovecot, etc, and running Fedora 7.
I'd like something using as little energy as possible, given that I need 200-300GB in disk-space. But I don't need a monitor (or graphics card).
Is anyone running such a system? If so, I'd be interested to hear the hardware details.
My home "server" is a Dell OptiPlex GX280 I got from a used computer place here in Austin. I want to say I paid $599 for the machine.
It came with a Pentium 4 3.0 GHz processor, two gigs of memory and an 80GB SATA drive. I bought two 500GB SATA drives to put in it - I think they were $129 each. The Intel video is built onto the motherboard so I didn't have to pay anything extra for it, ditto the gigabit ethernet NIC. Enough performance to handle about 20 pretty low volume web sites and the mail for each of those domains. Believe it or not, mail is a heck of a lot more load than the web traffic.
I like the OptiPlex machines because they are designed to be business class network workstations. NIC and video are provided and Just Work(TM) with Fedora or RHEL.
-
Bob Goodwin -
Craig White -
Craig White -
Ed K. -
Howard Howell -
Kam Leo -
Karl Larsen -
Lamar Owen -
thomas cameron -
Tim -
Timothy Murphy