t's a vastly underutilized tool for environments where you don't want users messing with the machines. I remember getting annoyed the first time I sat down at a box which wouldn't let me even look at the start menu.
In our undergrad labs at cs.usyd.edu.au, there was a low-end pentium for the sole purposes of ftping files from your floppy to your 3meg quota'd ugrad account on the nix machines. It was win3.1 (even though this was in 1998-2000), and all it _appeared_ to have was a crappy ftp client and 2 other semi-useless programs. You were given a 3 minute time-limit to use this machine. But one day, I recursively transferred the wrong files, and the ftp client was crap, and couldn't recursively remove directories, so I went to the c:\windows directory (or whatever), in the ftp client, selected command.com, and clicked the "run" button. I then was in a dos shell where I could deltree.
Moral of the story: There is no security in removing the start button:)
OK - debian seemed to have one version there - r5, whatever that is. How does it handle apt-get upgrades? If r5 is reffering to something like stable, then even stable changes over time (contrary to what some poeple think;-). So do they take the checksums from a machine that was just apt-get upgraded last night, or what? If they mean an actual yearly or half yearly release, who on this world does not apt-get upgrade when there is a security fix released? So your system sure as hell aint going to match theirs.
Then I can't imagine how you would be able to automate this, so it checks all the binaries in/bin/sbin,/usr/sbin etc - do they have some alternative to HTTP for their database?
Too bad we cant bill them back for my salary, and lost network resources, like we can do for un-requested faxes.
And arrest them for sending porn with out verifying a person's age. Around here, you would be either fined ( bookstore ) or arrested ( individual ) for trying such a stunt in 'real life'.
Hmm, interesting problem: If you come to Australia, and break a law, and then go back to America, we can requst to get you extradited to Australia, right (not that the American government would comply)?
Similarly, if Australia passes a law banning some forms of spam (as we saw earlier, we would have to kick out Loser Alston first), and you send spam, using *my* resources, break *our* laws, physically in *our* country (the electrons are passing through my server that is under my desk), then can we not request extradition?
But it's actually a hard job with the consumers wanting the impossible ("It must be our RIGHT to have FREE UNLIMITED broadband")
Exactly, who ever heard of something really useful that benefits the community, but expensive, being available for free. Like free roads. Or free education. Or free healthcare.
Yeah. Roads are fairly essential. Edumacation is pretty darn essential for a working society. Healthcare is essential (even if American's don't think so). I think broadband is in a different category. I certainly don't want to pay for your pr0n viewing pleasures with _my_ taxes.
What's the bandwidth of that trunk? Also, what's the capaity of the connections between each 16-port card and the backplane?
Just curious... suppose all the units on a 16-port card have 1Gbps each, but only 8Gb total to the backplane. Then the backplane, in turn, has only 8Gb to the other switch. These are just made up numbers, but how would beowulf handle it? Can it group jobs requiring higher communication throughput onto the nodes which are closer to eah other? Does it have to be told the topology, or does it figure it out?
Sounds like they have 10 ports @ 1Gbit each free, so get about 10Gbit between the switches. When we (position 180 on the latest top500) were investigating thin-tree connections, we thought that we might be able to effectively run one job on a third of the nodes, another on the second third, and miscellaneous jobs on the 3rd (since we have two or 3 researchers who like to use lots of resources, and people like me who only need a single proc at a time). So you just partition the nodes and only allow your mpi jobs to sit on one group at a time.
Then there are things like openmosix which deal with topology automatically somehow. They will try to calculate the speed of the interconnect and the various nodes' procs (if heterogenous), and work out the distribution in the most efficient way. I am going to try to convince my sysadmin to try out openmosix on the lesser-used nodes of the cluster, because there is a feature of it that I think one of us might like - the combination of memory of the different nodes in one big contigous space, but right now we are busy cleaning up after the upgrade.
Re:Sure, off the shelve cheap stuff...
on
Coolest Cluster Ever
·
· Score: 5, Informative
I really like the picture on the sites frontpage. i can imagine the small size of the Shuttles being an advantage, not to mention the "coolness" factor looking at it. (i assume the "cool" in the intro refers to emotion and not teperature!!)
But getting computation done cheap is no longer the challenge. It's getting the data from one node to the other. They still need "custom" expensive equipment for this.
I see they use 3com gigabit ethernet. having this 300+ gigabit switch capability is not "cheap".
Until one can buy this kind of networking equipment for really cheap, we shouldn't mention things like "of the shelve Beowulf super computer in the top 100".
Up until 3 hours ago, we said we were the only machine in the top500 dedicated solely to astrophysics. Now we are one of 2:)
But we use 180 processors, @ 2.15 GHz, and get about 0.90 Gflops/processor, whereas they get about 0.88 gflops/proc, getting us in the top 180th.
The difference being they have faster memory, and we have a big badass switch. They have two switches, with something like only 10 gigabit between the switches! We have 250gbit within our one switch. a third of our nodes have 2 gigs RAM, and we also have room for upgrade to more nodes on our switch, and they don't. So, in the words of Nelson "Hee haw!":)
When they say $1000 per proc, they are not factoriing in their two switches. This will bring the price up to about $500,000, unless someone is donating a switch or 2:)
We have about the same cost ratio - something like 250,000.au dollars for all our procs (and maybe switch - I don't know the details) - half the performance, half the cost.
These SIIA people are worse than the MPAA and RIAA combined. They are actively stealing MY money that I have ALREADY paided by squelching free dissemination of information. They are doing this purely as a way to gain market share for their members.
You don't live in Canada, do you?
Over there, I hear people pay taxes for blank cd's. I would personally be very pissed off if I had to pay $0.50 tax on a blank cd, given that I have never burned music onto CD's - I use them for research and backup.....I have already brought with my hard "earned" tax dollars whereas the MPAA and RIAA are only targeting potential costumers.
As of this weekend, my 8MB, 33MHz, 2Gig disk, 486 runs debian testing. Good for an NFS link, RAM usage hits 5MB into swap, so all the ssh connections (all 2 of them) sit in swap, and have to be swapped in. But nfs-kernel-server takes bugger all, and doesn't get swapped, because it is kernel, so is quite fast.
I forgot to mention, the way I do an apt-get install/upgrade/remove, etc is to run apt-get on my 650MHz laptop, in a chroot environment over nfs, with/proc and/dev/ just --bind mounts to the local system (so programs don't barf on not having permissions to write to device nodes, and not having/proc files):)
Much much faster, concidering dpkg likes to chew through 16Megs of RAM:)
I once did a kernel compile on this machine, years ago, when it was deadrat 4, and nevel will, again.
i run a modified version of redhat 4.2 on a cobalt qube. It's 150 mhz and it has 32 mb ram, and works as a great fileserver for 10 users. whoever said you need alot of computing power for a server is wrong
As of this weekend, my 8MB, 33MHz, 2Gig disk, 486 runs debian testing. Good for an NFS link, RAM usage hits 5MB into swap, so all the ssh connections (all 2 of them) sit in swap, and have to be swapped in. But nfs-kernel-server takes bugger all, and doesn't get swapped, because it is kernel, so is quite fast.
Did I mention that the hard disk has plenty of bad sectors? Haven't lost any data yet:)
1)Eolas wins, microsoft is crippled. 2)Microsoft wins, stupid patents are crippled.
Well, in response to 2 - it is not a stupid patent. This is a case where patent laws are a Good Thing - this patent was shown to Microsoft 2 years before M$ implmented the code, so it probably was "novel, and not obvious to someone versed in the art *at that time*".
And if M$ win, will it be because finally stupid patents have been realised to be bogus, or M$ has so much money to pay lawyers, that they just had to pay enough money to get the case struck down - or bought the judge or politicians?
I somewhat doubt it will be based on merit - but you probably know the US justice system more than me - fortunately, I only have to *hear* about the horrors....
Yep - I have one - he discovered this scourer thingy, and liked to throw that up in the air an chase it, and I discovered that I could throw it to him, he would run, jump, catch, throw, jump, catch, run, drop at my feet. And then whinge if I didn't repeat.
Back almost not off-topic, recently he discovered a toy koala that grips onto pencils, in my electronics room, and picked it up and played with it. I took it off him and put it back on the pencil, because I don't want it covered in cat slobber, so he went back to the pencil to pick it up again. Then I hid it in a jar, and he put his paws down the jar, and picked it out. Very skilled. So I hid it elsewhere, etc, etc, etc. Darn cat. Steals all my wire too. I lost some batteries, pens, and most annoyingly, a peice of plastic off a monitor I was fixing, then discovered him pawing under the fridge, so I looked under the fridge and found his stash. All, except for one button-battery! Probably ate the darn thing - he did stop eating for a week at one stage.....
If you are ten people, one of them could be fired, by your argument, without anybody noticing.
...
One has to do the cost/benefit thing carefully first
Arg, the joys of the commercial world;)
Nope, we are researchers, don't have to worry about money, except when the Howard Government wants to rip out all of our funding to fund the latest War on Terror(TM)
But for 99% of normal peoples taskes 10% whont matter.
10% never matters. We regularly run simulations here that take a month. What is 10% on top of a month? 3 days. If you have already been waiting 30 days, what does another 3 matter? It probably corresponds to the weekend anyway.....
Hmm, I'm tracking unstable, and keep periodicly trying apt-get install phoenix, but it still aint there. Is anyone working on this one? Of course, I am perfectly happy with galeon (I will never need to touch mozilla again. Yay!), but is perhaps phoenix even smaller?
"One of the things we're discovering is that people are not aware that that they are engaging in conduct which is clearly illegal," said RIAA lawyer Cary Sherman.
I like the clearly illegal comment which comes directly after saying that people are not aware that it is illegal.
That sounds like the scientific use of clearly.
Conclusion to my thesis -- "It is trivial to show that it is clearly obvious that this is not woofly."
The advantage of a multibeam setup is noise rejection. If you pick up the same signal on more than one reciever at once, you can be certain that's not the signal you are looking for.
If you pick it up on one, move the dish to put another beampath on the signal, and it's still there, it's a clear sign that you aren't looking at a terrestrial signal (or problems in the reciever.)
I notice that they will make use of the multibeam receiver to get effectively a 13 times increase in stuff they can look at at one time.
I wouldn't think this would be terribly useful. When you go up into the focus cabin, and realise the 13 recievers are separated by not more than about 40cm (the dish is 64 metres across) - ie, you are looking at an area of the sky about.5 degrees across (quick back of the envelope calculation there, so to speak). What in the sky is within.5 degrees? Globular clusters and the centre of the milyway galaxy (or indeed, other galaxies). You won't find life in globular clusters, because there are simply far too many stars too close together, and life would be cooked. Same thing for the centre of our our galaxy. And since you can only see entire globular clusters or supernova in other galaxies, I refuse to believe any civilasation could produce more radiation than a supernova, so we won't be able to see anything that far out!
Not to mention I really hate seeing such a useful instrument such as the multibeam receiver wasted on such a useless task as SETI, but they are probably (hopefully) only piggy backing on the electrons going to other experiments.:)
Translate amounts into some kind of real currency that doesn't involve kangaroos raping koalas. Something like: USD, CAD, GBP, EUR, or SFR.
I know what an AU$ is, I live here. I know how to convert to USD and EUR. Never heard of CAD, but guess Canadian, don't know how to convert. GBP? Great Britain Pounds? SFR? Star Formation Rate? What does that have to do with currency?
I don't mean to be cruel, for I could use the same advice, but if that lady in the hospital photo lost a some weight, it looks like the stone would have *missed* her rather large tummy.
The physics are simple: Bigger people are bigger targets.
In high energy physics, a particle with a lot of mass tends to have more of something with call cross-section, analagous to a size you give a physical ball. So you are saying this woman is a high energy particle? Can I throw neutrons at her? Will they bounce back and hit me in the face, just like in Thompson scattering?
Mozilla took two hours on a system like that? When was this, Milestone 6?
About then;)
I gave up trying to build anymore, and just waited for the debian packages:)
Re:Sigh... I want a *cooler* processor...
on
Pentium 4 2.8GHz
·
· Score: 2
A complete computer system produces quite a lot of heat. There's the processor, graphics card and the harddrive. The power supply is rarely efficient and does its share to heat the room. The worst offender however is probably the monitor. Overall, there's much more than 60W coming from your computer system.
True. Almost. The power supply is quite efficeient. Switch mode power supplies are typically 95 percent efficient. But yes, the rest does pump out a bit of heat. Interestingly enough, since almost all of the energy going into your computer comes out as heat (the monitor puts out a bit of light as an added bonus, and the fans make a bit of noise, but this noise gets turned into heat if it doesn't escape the enclosure, and is bugger all, in terms of energy, anyway), and airconditioners (by the second law of thermodynamics) will spend the same amount of energy plus some to extract that heat, if you are using your air-conditioner, then you have to more than double the power used by your computer! Quite a concideration when you build your next beowulf (as we are - I don't know whether we have the extra air conditioning installed yet, that our new nodes require).
Slashdot Mirror
t's a vastly underutilized tool for environments where you don't want users messing with the machines. I remember getting annoyed the first time I sat down at a box which wouldn't let me even look at the start menu.
:)
In our undergrad labs at cs.usyd.edu.au, there was a low-end pentium for the sole purposes of ftping files from your floppy to your 3meg quota'd ugrad account on the nix machines. It was win3.1 (even though this was in 1998-2000), and all it _appeared_ to have was a crappy ftp client and 2 other semi-useless programs. You were given a 3 minute time-limit to use this machine. But one day, I recursively transferred the wrong files, and the ftp client was crap, and couldn't recursively remove directories, so I went to the c:\windows directory (or whatever), in the ftp client, selected command.com, and clicked the "run" button. I then was in a dos shell where I could deltree.
Moral of the story: There is no security in removing the start button
Oh my god.. this one is incredibly bitching! It even tries to get your social security #, maiden name, bank account # and pin .. etc
Heh heh. I just filled in some bogus info (had to guess what US phone numbers look like, etc). Is SSN 9 digits long?
But concider it evolution. Those stupid enough to give their SSN, passwds, pin, account numbers etc deserve to lose.
OK - debian seemed to have one version there - r5, whatever that is. How does it handle apt-get upgrades? If r5 is reffering to something like stable, then even stable changes over time (contrary to what some poeple think ;-). So do they take the checksums from a machine that was just apt-get upgraded last night, or what? If they mean an actual yearly or half yearly release, who on this world does not apt-get upgrade when there is a security fix released? So your system sure as hell aint going to match theirs.
/bin /sbin, /usr/sbin etc - do they have some alternative to HTTP for their database?
Then I can't imagine how you would be able to automate this, so it checks all the binaries in
Doesn't seem overly useful to me....
Too bad we cant bill them back for my salary, and lost network resources, like we can do for un-requested faxes.
And arrest them for sending porn with out verifying a person's age. Around here, you would be either fined ( bookstore ) or arrested ( individual ) for trying such a stunt in 'real life'.
Hmm, interesting problem: If you come to Australia, and break a law, and then go back to America, we can requst to get you extradited to Australia, right (not that the American government would comply)?
Similarly, if Australia passes a law banning some forms of spam (as we saw earlier, we would have to kick out Loser Alston first), and you send spam, using *my* resources, break *our* laws, physically in *our* country (the electrons are passing through my server that is under my desk), then can we not request extradition?
Bring back hanging, I say.
Yeah. Roads are fairly essential. Edumacation is pretty darn essential for a working society. Healthcare is essential (even if American's don't think so). I think broadband is in a different category. I certainly don't want to pay for your pr0n viewing pleasures with _my_ taxes.
What's the bandwidth of that trunk? Also, what's the capaity of the connections between each 16-port card and the backplane?
Just curious... suppose all the units on a 16-port card have 1Gbps each, but only 8Gb total to the backplane. Then the backplane, in turn, has only 8Gb to the other switch. These are just made up numbers, but how would beowulf handle it? Can it group jobs requiring higher communication throughput onto the nodes which are closer to eah other? Does it have to be told the topology, or does it figure it out?
Sounds like they have 10 ports @ 1Gbit each free, so get about 10Gbit between the switches. When we (position 180 on the latest top500) were investigating thin-tree connections, we thought that we might be able to effectively run one job on a third of the nodes, another on the second third, and miscellaneous jobs on the 3rd (since we have two or 3 researchers who like to use lots of resources, and people like me who only need a single proc at a time). So you just partition the nodes and only allow your mpi jobs to sit on one group at a time.
Then there are things like openmosix which deal with topology automatically somehow. They will try to calculate the speed of the interconnect and the various nodes' procs (if heterogenous), and work out the distribution in the most efficient way. I am going to try to convince my sysadmin to try out openmosix on the lesser-used nodes of the cluster, because there is a feature of it that I think one of us might like - the combination of memory of the different nodes in one big contigous space, but right now we are busy cleaning up after the upgrade.
I really like the picture on the sites frontpage.
:)
:)
:)
.au dollars for all our procs (and maybe switch - I don't know the details) - half the performance, half the cost.
i can imagine the small size of the Shuttles being an advantage, not to mention the "coolness" factor looking at it. (i assume the "cool" in the intro refers to emotion and not teperature!!)
But getting computation done cheap is no longer the challenge. It's getting the data from one node to the other. They still need "custom" expensive equipment for this.
I see they use 3com gigabit ethernet. having this 300+ gigabit switch capability is not "cheap".
Until one can buy this kind of networking equipment for really cheap, we shouldn't mention things like "of the shelve Beowulf super computer in the top 100".
Up until 3 hours ago, we said we were the only machine in the top500 dedicated solely to astrophysics. Now we are one of 2
But we use 180 processors, @ 2.15 GHz, and get about 0.90 Gflops/processor, whereas they get about 0.88 gflops/proc, getting us in the top 180th.
The difference being they have faster memory, and we have a big badass switch. They have two switches, with something like only 10 gigabit between the switches! We have 250gbit within our one switch. a third of our nodes have 2 gigs RAM, and we also have room for upgrade to more nodes on our switch, and they don't. So, in the words of Nelson "Hee haw!"
When they say $1000 per proc, they are not factoriing in their two switches. This will bring the price up to about $500,000, unless someone is donating a switch or 2
We have about the same cost ratio - something like 250,000
It should be
:)
Imagine a Mosix cluster of those
Um, openmosix, you corporate scum
This time, from where I work, yet another 2002 eclipse website
They plan on having a live braodcast, but somehow I don't think that will work.
These SIIA people are worse than the MPAA and RIAA combined. They are actively stealing MY money that I have ALREADY paided by squelching free dissemination of information. They are doing this purely as a way to gain market share for their members.
....I have already brought with my hard "earned" tax dollars whereas the MPAA and RIAA are only targeting potential costumers.
You don't live in Canada, do you?
Over there, I hear people pay taxes for blank cd's. I would personally be very pissed off if I had to pay $0.50 tax on a blank cd, given that I have never burned music onto CD's - I use them for research and backup.
Nice tyopo.
As of this weekend, my 8MB, 33MHz, 2Gig disk, 486 runs debian testing. Good for an NFS link, RAM usage hits 5MB into swap, so all the ssh connections (all 2 of them) sit in swap, and have to be swapped in. But nfs-kernel-server takes bugger all, and doesn't get swapped, because it is kernel, so is quite fast.
/proc and /dev/ just --bind mounts to the local system (so programs don't barf on not having permissions to write to device nodes, and not having /proc files) :)
:)
I forgot to mention, the way I do an apt-get install/upgrade/remove, etc is to run apt-get on my 650MHz laptop, in a chroot environment over nfs, with
Much much faster, concidering dpkg likes to chew through 16Megs of RAM
I once did a kernel compile on this machine, years ago, when it was deadrat 4, and nevel will, again.
i run a modified version of redhat 4.2 on a cobalt qube. It's 150 mhz and it has 32 mb ram, and works as a great fileserver for 10 users. whoever said you need alot of computing power for a server is wrong
:)
As of this weekend, my 8MB, 33MHz, 2Gig disk, 486 runs debian testing. Good for an NFS link, RAM usage hits 5MB into swap, so all the ssh connections (all 2 of them) sit in swap, and have to be swapped in. But nfs-kernel-server takes bugger all, and doesn't get swapped, because it is kernel, so is quite fast.
Did I mention that the hard disk has plenty of bad sectors? Haven't lost any data yet
1)Eolas wins, microsoft is crippled.
2)Microsoft wins, stupid patents are crippled.
Well, in response to 2 - it is not a stupid patent. This is a case where patent laws are a Good Thing - this patent was shown to Microsoft 2 years before M$ implmented the code, so it probably was "novel, and not obvious to someone versed in the art *at that time*".
And if M$ win, will it be because finally stupid patents have been realised to be bogus, or M$ has so much money to pay lawyers, that they just had to pay enough money to get the case struck down - or bought the judge or politicians?
I somewhat doubt it will be based on merit - but you probably know the US justice system more than me - fortunately, I only have to *hear* about the horrors....
Anyone? Hello? Fetching cats?
Yep - I have one - he discovered this scourer thingy, and liked to throw that up in the air an chase it, and I discovered that I could throw it to him, he would run, jump, catch, throw, jump, catch, run, drop at my feet. And then whinge if I didn't repeat.
Back almost not off-topic, recently he discovered a toy koala that grips onto pencils, in my electronics room, and picked it up and played with it. I took it off him and put it back on the pencil, because I don't want it covered in cat slobber, so he went back to the pencil to pick it up again. Then I hid it in a jar, and he put his paws down the jar, and picked it out. Very skilled. So I hid it elsewhere, etc, etc, etc. Darn cat. Steals all my wire too. I lost some batteries, pens, and most annoyingly, a peice of plastic off a monitor I was fixing, then discovered him pawing under the fridge, so I looked under the fridge and found his stash. All, except for one button-battery! Probably ate the darn thing - he did stop eating for a week at one stage.....
If you are ten people, one of them could be fired, by your argument, without anybody noticing.
...
;)
One has to do the cost/benefit thing carefully first
Arg, the joys of the commercial world
Nope, we are researchers, don't have to worry about money, except when the Howard Government wants to rip out all of our funding to fund the latest War on Terror(TM)
For straight CPU intensive tasks it matters.
But for 99% of normal peoples taskes 10% whont matter.
10% never matters. We regularly run simulations here that take a month. What is 10% on top of a month? 3 days. If you have already been waiting 30 days, what does another 3 matter? It probably corresponds to the weekend anyway.....
Hmm, I'm tracking unstable, and keep periodicly trying apt-get install phoenix, but it still aint there. Is anyone working on this one? Of course, I am perfectly happy with galeon (I will never need to touch mozilla again. Yay!), but is perhaps phoenix even smaller?
That sounds like the scientific use of clearly.
Conclusion to my thesis -- "It is trivial to show that it is clearly obvious that this is not woofly."
The advantage of a multibeam setup is noise rejection. If you pick up the same signal on more than one reciever at once, you can be certain that's not the signal you are looking for.
:)
If you pick it up on one, move the dish to put another beampath on the signal, and it's still there, it's a clear sign that you aren't looking at a terrestrial signal (or problems in the reciever.)
Duh. Didn't think of that one - thanks
I notice that they will make use of the multibeam receiver to get effectively a 13 times increase in stuff they can look at at one time.
.5 degrees across (quick back of the envelope calculation there, so to speak). What in the sky is within .5 degrees? Globular clusters and the centre of the milyway galaxy (or indeed, other galaxies). You won't find life in globular clusters, because there are simply far too many stars too close together, and life would be cooked. Same thing for the centre of our our galaxy. And since you can only see entire globular clusters or supernova in other galaxies, I refuse to believe any civilasation could produce more radiation than a supernova, so we won't be able to see anything that far out!
:)
I wouldn't think this would be terribly useful. When you go up into the focus cabin, and realise the 13 recievers are separated by not more than about 40cm (the dish is 64 metres across) - ie, you are looking at an area of the sky about
Not to mention I really hate seeing such a useful instrument such as the multibeam receiver wasted on such a useless task as SETI, but they are probably (hopefully) only piggy backing on the electrons going to other experiments.
...have any idea of how expensive 800 AUD is?
Translate amounts into some kind of real currency that doesn't involve kangaroos raping koalas. Something like: USD, CAD, GBP, EUR, or SFR.
I know what an AU$ is, I live here. I know how to convert to USD and EUR. Never heard of CAD, but guess Canadian, don't know how to convert. GBP? Great Britain Pounds? SFR? Star Formation Rate? What does that have to do with currency?
I don't mean to be cruel, for I could use the same advice, but if that lady in the hospital photo lost a some weight, it looks like the stone would have *missed* her rather large tummy.
The physics are simple: Bigger people are bigger targets.
In high energy physics, a particle with a lot of mass tends to have more of something with call cross-section, analagous to a size you give a physical ball. So you are saying this woman is a high energy particle? Can I throw neutrons at her? Will they bounce back and hit me in the face, just like in Thompson scattering?
Give it a rest, people.
;)
*PLONK*
Wow! There is a killfile function on slashdot?! Show me where it is!
Mozilla took two hours on a system like that? When was this, Milestone 6?
;)
:)
About then
I gave up trying to build anymore, and just waited for the debian packages
A complete computer system produces quite a lot of heat. There's the processor, graphics card and the harddrive. The power supply is rarely efficient and does its share to heat the room. The worst offender however is probably the monitor. Overall, there's much more than 60W coming from your computer system.
True. Almost. The power supply is quite efficeient. Switch mode power supplies are typically 95 percent efficient. But yes, the rest does pump out a bit of heat. Interestingly enough, since almost all of the energy going into your computer comes out as heat (the monitor puts out a bit of light as an added bonus, and the fans make a bit of noise, but this noise gets turned into heat if it doesn't escape the enclosure, and is bugger all, in terms of energy, anyway), and airconditioners (by the second law of thermodynamics) will spend the same amount of energy plus some to extract that heat, if you are using your air-conditioner, then you have to more than double the power used by your computer! Quite a concideration when you build your next beowulf (as we are - I don't know whether we have the extra air conditioning installed yet, that our new nodes require).