current x potential x power_factor = power
Power factor varies from 0 (pure reactance) to 1 (pure resistance) and is equal to the cosine of the difference between the current phase and voltage phase. Other than that minor goof, a very nice write-up.
The most compelling reason for using the "Kill-A-Watt" over a multimeter is safety. We had someone at work who wanted to brew up a power line wattmeter, and I persuaded him that it would be cheaper and much safer to buy a ready made wattmeter. The project did get mentioned on Slashdot a few months back - adapting a Mac Mini to run on batteries.
That was, what, 5-6 years ago? Clearly they missed an excellent opportunity to expand gcc then, but that would compete with their internal C compiler. Which ought to be killed off due to the stupid license manager which hinders every single compilation (so much so, that the Solaris O.S. buildmeisters have turned it off).
I just downloaded Sun's Studio 11 package, no license manager in sight (which, iirc, stopped with Workshop 6 Update 2). The software is available for a free download, but support will cost you (which makes sense). Sun's compilers have usually given much better performance than gcc on Sun hardware and appear to give better performance on the Opteron as well.
BTW, Sun's first 64 bit compiler shipped about the same time as Solaris 7, late 1998.
Free versions of Solaris, free compilers and competively priced Opteron workstations...
Pulling a number out of the air - I'd guess that one of T1 boxes would easily keep up with a dozen U5's - a single T1 would have at most 2x the power consumption of a single U5. The bang for buck of the bank of U5's is better than a T1 provided that your power and airconditioning costs are low, space is not at a premium, your application works nicely in a cluster and most importantly: you have adequate RAM on the U5's. The DDR-2 memory for the T1's is a lot cheaper than the EDO ram used by the U5 (I know - I have a U10, which is the same motherboard).
As far as geographic diversity - by the time that you add up the costs of setting up even a minimalist data center - those costs will far outweigh the difference between the T1 and the U5 cluster. One exception would be a distributed multiprocessor system running in home networks.
I think you're being optimistic about the cache for the multicore PPC 440.
As far as power requirements - once you fill up all 16 memory slots, the memory will be consuming more power than the CPU. That implies that you wouldn't see much of an energy savings with the multicore PPC 440. To get further energy reduction would require more attention to reducing the memory I/O - maybe better cache design, maybe better software.
If I had it my way, everyone would use a proper Sparc box for their servers.
Agreed.
One of the promises of Sun's Niagara servers is greatly increased performance per watt - which is one of the themes of the article.
Something similar has come for a project at work - there's an Apple Xserve cluster that will be installed in a few days - one of the big items in preparation was finding a spot with the necessary power and A/C capacity - and this is small potatoes compared to a typical hosting site. While waiting for the Apple cluster to be set up, I'm making do with a cluster of Sun V-120's at the "software provider". In both cases it is much easier running a remote session with ssh than the equivalent with Windows.
One of the things that you would want to do is have galvanic isolation between the battery bank and solar panels - an isolated DC/DC converter will do nicely and allow for optimum power production from the solar cells.
The big power eaters in your house - refridgerator, air conditioners, etc, would still require A/C as A/C motors are much more efficient.
AC motors are cheaper than DC motors, but not necessarily more efficient.
The trend with refrigeration equipment is to go with variable speed motors which use an inverter to convert DC to variable frequency AC. The DC voltages are higher than the ones talked about for server rooms, so it would to be treated similarly to AC wiring. It would make sense if the power was provided by a renewable energy source - the DC mains could then be tied directly to the batteries (with appropriate current limiters).
A better solution would be to replace the wall warts with mini switching power
It would be really nice if there was a standard DC voltage portable devices (13.6 would be an obvious choice) and the devices handled the conversion down to the point of use voltages. I would love to have a single brick (maybe with a bit of battery back up) for all the devices with wall warts. (ISTR this was an "Ask Slashdot" topic.) An example of a properly doe part are the SoeKris computers - they will run on 6-15VDC.
The 50hz was chosen because that was the max possible at the time, and 60hz was chosen because tech improved to that point
One story I've read is that the Europeans decided to standadize on 50 Hz, and the Americans chose 60 Hz to avoid competition. This is somewhat suspect because the US had standardized on 115V by virtue of Edison's Pearl Street installation and commercial AC frequencies varied from 25 to 133 Hz. The last 50Hz installation was phased out in 1948 and DC lasted until 2000.
The real reason for 50/60Hz is that is a good compromise between the needs of long distance power transmission (max power flow inversely proportional to frequency) and size of transformers. 25 Hz power was common in the US where the load involved a commutator, such as series motors for AC railroad electrification, and synchronous (rotary) converters for DC power (aluminum smelting, electric railroads).
Arguably, power should be 3-phase down to the point where it's rectified to DC, because 3-phase rectifiers need far less filtering, but nobody does this for small loads.
Especially if you can swing a 12-pulse rectifier - which gives much smoother DC and less harmonics on the AC side. This is becoming less of an issue with PFC SMPS's.
Typically, when you get to the computer, there's a conversion from the line voltage (120-240VAC, 48VDC, etc) to internal distribution voltages of 5-12VDC, then another conversion and regulation just before each device, usually downward to something like 3.3VDC.
The big reason for multiple conversions is that the latest and greatest chips use ~100W of power at a bit over 1V - so you 100A from the power supply - there is no way that you can transmit that much power at that low a voltage for more than a few millimeters with standard PC board traces. The solution is to use a buck converter right next to the CPU - and these converters can be very efficient (they have to be to stay cheap).
What a lot of people are missing the baot on wrt DC/DC vs AC/DC conversion is that it is easier to get high efficiency with low voltage parts. The Rdson of MOSFET's rises rapidly when the Vds rating goes above 60V, silicon Schottky diodes are only good up to 200V (although SiC Schottkys are starting to show up). The fact that the input voltage is more or less constant means a lot less filtering and no need for PFC.
If the legislature and courts take the view that copyright is akin to property...
If "copyright" is property, then why aren't the governments collecting property tax? If it is property, then can we expect the "copyright" owners to be liable for their property?
...but in some ways it even goes beyond incompetence, to what almost seemed like a willful destruction of the company by Richard E. Belluzzo.
Before he laid waste to SGI, Belluzo was responsible for HP scaling back on the 9000/700 series desktops (HP-UX). In the early 90's, HP had Lotus 123 and Ami-Pro ported to run on HP-UX and VUE was a much nicer looking windowing system than Windblows 3.1. Belluzo convinced HP's management to drop efforts on HP-UX for the desktop and move to Win NT (which really didn't fulfill the promise until Win2k).
Time will tell if that's true, but the massive increase in power density is great news for hybrids. To really make hybrids shine, you need to do as much regenerative braking as possible - which means capturing braking energy at high speeds - which means very high power requirements. As you said, the good news is that if the drivetrain is good for near panic stops in regenerative braking, then the accelaration should be spectacular.
It will be an interesting race between ultracapacitors and these batteries - the ultracaps have the edge in power density (already 3kW/kg) and cycle time 10E5 or more charge/discharge cycles - but Li-ion has the energy density.
A fairly insightful point. There's a lot of energy savings to be had by requiring higher albedos for roofing material used in all but the coldest spots in the U.S. Better insulation would go a long way as well as well as more thermal mass to allow more efficient heating and cooling.
The prevalent theory on Slashdot is that Ma Bell gets the infrastructure paid for by government subsidy,
There is a government subsidy involved called eminent domain. Ma Bell and other utilities get a very substantial break in the cost of their right-of-way. Part of the trade-off is that the utility in question needs to operate in the public interest, convenience or necessity.
I have nothing against telco's making money (being a Verizon shareholder), but the telco's do need to hold up their part of the bargain with respect to property taken by eminent domain.
Next, we should be building the advanced Integral Fast Reactors (IFR's) which Argonne Labs designed by about 1994. The program was shut down by Clinton.
Hmmm. One of the books in my collection is The EBR-II Fuel Cycle Story copyrighted in 1987 (an ANS punlication). What I've herd of the IFR design sounds like an update of the EBR-II. Seem to recall that Argonne was in charge of INEL.
Heard some stories about the efforts to shut down the program...
Wondering if an HTGR design may be better suited for providing the process heat needed for the tar sands (IIRC, the CANDU may be good for 550F max).
First off, the electron velocity in wire is much less than the propagation velocity through the same wire.
Now for the fun part - What is the velocity of propagation?
For frequencies were the inductive reactance of the conductor is significantly larger than the resistance of that conductor at that frequency (think skin effect), then the velocity of propagation is c divided by the square root of the effective relative dielectric constant. This is often referred to as an LC transmission line since propagation is dominated by the series inductance and shunt capaitance. LC lines have a propagation velocity independent of frequency (at least to the first order). As an example, coaxial cable with a solid polyethylene dielectric will have a propagation velocity of 0.66c, which would be valid from a few hundred kHz to several GHz.
When the the conductor resistance is greater than the inductive reactance, then the line becomes an RC line where the "propagation velocity" is dependent on frequency (dispersive) and the time for a transition to propagate is proportional to the square of the line length. The effective "propagation velocity" is going to be a lot less than c. Turns out that the interconnects on chips are RC lines - and it is often necessary to insert inverters on a line to speed things up (recall that propagation time varies with the square of the line length) - a good rule of thumb is to space the inverters so the the propagation delay equals the gate delay.
The RC problem is why loading coils were put on phone lines - the inductive reactance of the coils is larger than the resistance and the line becomes an LC. The loading coils are bad news for DSL - and an unloaded line looks like an LC line at the frequencies used by the DSL modems.
A good reference for this is High Speed Digital Design, a Handbook of Black Magic by Johnson and Graham.
Why are they made in pieces and shipped to Florida? Jobs in Utah.
Utah has a much drier climate than Florida - which helps in handling the ammonium perchlorate oxidizer - and is much closer to the AP production site in Henderson, NV. In addition, the folks in Utah have the production facilities to make SRM's and the ATK-Thiokol plant is in the middle of nowhere - fewer people to worry about if something really bad happens.
If they had been built at Cape Canaveral they'd be in one piece,
Casting a million pounds of fuel at once???? Yuk!
With something that big, you want to make it in sections to make sure that the process is done right - and the timing for the process is fairly critical - once you've applied the bond-liner, you've got a few hour window starting 24 to 48 hours later to complete the casting - that's assuming that the bond-liner was applied correctly and assuming that the grain does not have any voids.
Our ability to extract coal is entirely dependent upon cheap oil (makes/powers the mining and transportation equipment)
Most of the really large equipment (think shovels that can pick up 50 m^3 at a time) at coal mines are electrically powered and it wouldn't be that difficult to convert the long distance haulage to electricity as well.
What's more of an issue for home is passing the decline curve for natural gas - but as the price of gas rises enough, extracting methane from hydrates is likely to become economical - and there is an incredible amount of methane in the form of hydrates.
Since the OSS "office" programs are moving to a common file format (.odt), having a healthy competition between the various offerings may end helping all of them. The more people who have a reason to switch away from M$-Orifice to an ODT application, the better the "market" is for all ODT applications. I'd really like to see the word processor "market" evolve to where the text editor "market" has been for the last couple of decades where there is still real choice in editors (my fave being NEdit).
Perhaps the worst aspect of MS's monopoly is the lack of effective competition to spur real innovation and product improvement. Note, for example, the almost complete lack of improvement in Internet Exploder between the decline of Netscape and the rise of Firefox.
Slashdot Mirror
Power factor varies from 0 (pure reactance) to 1 (pure resistance) and is equal to the cosine of the difference between the current phase and voltage phase. Other than that minor goof, a very nice write-up.
The most compelling reason for using the "Kill-A-Watt" over a multimeter is safety. We had someone at work who wanted to brew up a power line wattmeter, and I persuaded him that it would be cheaper and much safer to buy a ready made wattmeter. The project did get mentioned on Slashdot a few months back - adapting a Mac Mini to run on batteries.
Now if they were talking about "awk" on Solaris...
I just downloaded Sun's Studio 11 package, no license manager in sight (which, iirc, stopped with Workshop 6 Update 2). The software is available for a free download, but support will cost you (which makes sense). Sun's compilers have usually given much better performance than gcc on Sun hardware and appear to give better performance on the Opteron as well.
BTW, Sun's first 64 bit compiler shipped about the same time as Solaris 7, late 1998.
Free versions of Solaris, free compilers and competively priced Opteron workstations...
As far as geographic diversity - by the time that you add up the costs of setting up even a minimalist data center - those costs will far outweigh the difference between the T1 and the U5 cluster. One exception would be a distributed multiprocessor system running in home networks.
As far as power requirements - once you fill up all 16 memory slots, the memory will be consuming more power than the CPU. That implies that you wouldn't see much of an energy savings with the multicore PPC 440. To get further energy reduction would require more attention to reducing the memory I/O - maybe better cache design, maybe better software.
Agreed.
One of the promises of Sun's Niagara servers is greatly increased performance per watt - which is one of the themes of the article.
Something similar has come for a project at work - there's an Apple Xserve cluster that will be installed in a few days - one of the big items in preparation was finding a spot with the necessary power and A/C capacity - and this is small potatoes compared to a typical hosting site. While waiting for the Apple cluster to be set up, I'm making do with a cluster of Sun V-120's at the "software provider". In both cases it is much easier running a remote session with ssh than the equivalent with Windows.
The is less of an electrocution risk at HF than 60Hz - skin effect. On the other hand, RF burns can be pretty painful - and I speak from experience.
One of the things that you would want to do is have galvanic isolation between the battery bank and solar panels - an isolated DC/DC converter will do nicely and allow for optimum power production from the solar cells.
To paraphrase Rich Teer - who in their right mind puts a frame buffer on a server?
AC motors are cheaper than DC motors, but not necessarily more efficient.
The trend with refrigeration equipment is to go with variable speed motors which use an inverter to convert DC to variable frequency AC. The DC voltages are higher than the ones talked about for server rooms, so it would to be treated similarly to AC wiring. It would make sense if the power was provided by a renewable energy source - the DC mains could then be tied directly to the batteries (with appropriate current limiters).
It would be really nice if there was a standard DC voltage portable devices (13.6 would be an obvious choice) and the devices handled the conversion down to the point of use voltages. I would love to have a single brick (maybe with a bit of battery back up) for all the devices with wall warts. (ISTR this was an "Ask Slashdot" topic.) An example of a properly doe part are the SoeKris computers - they will run on 6-15VDC.
One story I've read is that the Europeans decided to standadize on 50 Hz, and the Americans chose 60 Hz to avoid competition. This is somewhat suspect because the US had standardized on 115V by virtue of Edison's Pearl Street installation and commercial AC frequencies varied from 25 to 133 Hz. The last 50Hz installation was phased out in 1948 and DC lasted until 2000.
The real reason for 50/60Hz is that is a good compromise between the needs of long distance power transmission (max power flow inversely proportional to frequency) and size of transformers. 25 Hz power was common in the US where the load involved a commutator, such as series motors for AC railroad electrification, and synchronous (rotary) converters for DC power (aluminum smelting, electric railroads).
Especially if you can swing a 12-pulse rectifier - which gives much smoother DC and less harmonics on the AC side. This is becoming less of an issue with PFC SMPS's.
Typically, when you get to the computer, there's a conversion from the line voltage (120-240VAC, 48VDC, etc) to internal distribution voltages of 5-12VDC, then another conversion and regulation just before each device, usually downward to something like 3.3VDC.
The big reason for multiple conversions is that the latest and greatest chips use ~100W of power at a bit over 1V - so you 100A from the power supply - there is no way that you can transmit that much power at that low a voltage for more than a few millimeters with standard PC board traces. The solution is to use a buck converter right next to the CPU - and these converters can be very efficient (they have to be to stay cheap).
What a lot of people are missing the baot on wrt DC/DC vs AC/DC conversion is that it is easier to get high efficiency with low voltage parts. The Rdson of MOSFET's rises rapidly when the Vds rating goes above 60V, silicon Schottky diodes are only good up to 200V (although SiC Schottkys are starting to show up). The fact that the input voltage is more or less constant means a lot less filtering and no need for PFC.
If "copyright" is property, then why aren't the governments collecting property tax? If it is property, then can we expect the "copyright" owners to be liable for their property?
Before he laid waste to SGI, Belluzo was responsible for HP scaling back on the 9000/700 series desktops (HP-UX). In the early 90's, HP had Lotus 123 and Ami-Pro ported to run on HP-UX and VUE was a much nicer looking windowing system than Windblows 3.1. Belluzo convinced HP's management to drop efforts on HP-UX for the desktop and move to Win NT (which really didn't fulfill the promise until Win2k).
Time will tell if that's true, but the massive increase in power density is great news for hybrids. To really make hybrids shine, you need to do as much regenerative braking as possible - which means capturing braking energy at high speeds - which means very high power requirements. As you said, the good news is that if the drivetrain is good for near panic stops in regenerative braking, then the accelaration should be spectacular.
It will be an interesting race between ultracapacitors and these batteries - the ultracaps have the edge in power density (already 3kW/kg) and cycle time 10E5 or more charge/discharge cycles - but Li-ion has the energy density.
A fairly insightful point. There's a lot of energy savings to be had by requiring higher albedos for roofing material used in all but the coldest spots in the U.S. Better insulation would go a long way as well as well as more thermal mass to allow more efficient heating and cooling.
There is a government subsidy involved called eminent domain. Ma Bell and other utilities get a very substantial break in the cost of their right-of-way. Part of the trade-off is that the utility in question needs to operate in the public interest, convenience or necessity.
I have nothing against telco's making money (being a Verizon shareholder), but the telco's do need to hold up their part of the bargain with respect to property taken by eminent domain.
Hmmm. One of the books in my collection is The EBR-II Fuel Cycle Story copyrighted in 1987 (an ANS punlication). What I've herd of the IFR design sounds like an update of the EBR-II. Seem to recall that Argonne was in charge of INEL.
Heard some stories about the efforts to shut down the program...
Wondering if an HTGR design may be better suited for providing the process heat needed for the tar sands (IIRC, the CANDU may be good for 550F max).
Now for the fun part - What is the velocity of propagation?
For frequencies were the inductive reactance of the conductor is significantly larger than the resistance of that conductor at that frequency (think skin effect), then the velocity of propagation is c divided by the square root of the effective relative dielectric constant. This is often referred to as an LC transmission line since propagation is dominated by the series inductance and shunt capaitance. LC lines have a propagation velocity independent of frequency (at least to the first order). As an example, coaxial cable with a solid polyethylene dielectric will have a propagation velocity of 0.66c, which would be valid from a few hundred kHz to several GHz.
When the the conductor resistance is greater than the inductive reactance, then the line becomes an RC line where the "propagation velocity" is dependent on frequency (dispersive) and the time for a transition to propagate is proportional to the square of the line length. The effective "propagation velocity" is going to be a lot less than c. Turns out that the interconnects on chips are RC lines - and it is often necessary to insert inverters on a line to speed things up (recall that propagation time varies with the square of the line length) - a good rule of thumb is to space the inverters so the the propagation delay equals the gate delay.
The RC problem is why loading coils were put on phone lines - the inductive reactance of the coils is larger than the resistance and the line becomes an LC. The loading coils are bad news for DSL - and an unloaded line looks like an LC line at the frequencies used by the DSL modems.
A good reference for this is High Speed Digital Design, a Handbook of Black Magic by Johnson and Graham.
Utah has a much drier climate than Florida - which helps in handling the ammonium perchlorate oxidizer - and is much closer to the AP production site in Henderson, NV. In addition, the folks in Utah have the production facilities to make SRM's and the ATK-Thiokol plant is in the middle of nowhere - fewer people to worry about if something really bad happens.
If they had been built at Cape Canaveral they'd be in one piece,
Casting a million pounds of fuel at once???? Yuk!
With something that big, you want to make it in sections to make sure that the process is done right - and the timing for the process is fairly critical - once you've applied the bond-liner, you've got a few hour window starting 24 to 48 hours later to complete the casting - that's assuming that the bond-liner was applied correctly and assuming that the grain does not have any voids.
And quite spectacularly from what I've heard - heard that is SOP for disposing of the old Poseidon motors.
The detonations can be a bit more spectacular if you use Acetylozone instead of the normal AP/Al solid fuel mix. ;-^)
Most of the really large equipment (think shovels that can pick up 50 m^3 at a time) at coal mines are electrically powered and it wouldn't be that difficult to convert the long distance haulage to electricity as well.
What's more of an issue for home is passing the decline curve for natural gas - but as the price of gas rises enough, extracting methane from hydrates is likely to become economical - and there is an incredible amount of methane in the form of hydrates.
Conversion factor: 1 KWH = 3412 BTU
Multiplying the average (not peak) wattage by 3.4 will get the BTU's per hour of cooling capacity.
Perhaps the worst aspect of MS's monopoly is the lack of effective competition to spur real innovation and product improvement. Note, for example, the almost complete lack of improvement in Internet Exploder between the decline of Netscape and the rise of Firefox.