You know, I used to be on the upgrade treadmill. I've got to get a new computer in order to play games etc. I've done two things that have changed the way I look at computers. First, I bought a Mac (Why? Because it is a great laptop for a engineer/programmer/student, and generally anyone. Besides the point anyway) I kept my PC, but it quickly stopped being used. Which brings me to my second point. I stopped playing PC games. Generally I just didn't have time, and then I didn't want to. I have a couple consoles that I use on rare occasions. Very rare occasions.
Now that I don't play PC games, there's no reason for me to own a PC. My powerbook does everything and more.
Anyone can use whatever they want, and I don't care, but I won't be going back the way of MS unless there's something that will make me want to switch. That doesn't seem to be happening yet.
What is the power draw at 100% load on a single computer as compared to the same computing power on multiple computers at lower loads?
In other words, how many CPUs are needed at 10% load to achieve the same computing power as a single CPU running at 100% load, and how does the power draw compare?
Now, if you were to run the same computation on one computer at lower load levels (say let something run overnight if available) how does the power draw compare?
I have a hypothosis: Power draw increases with the number of operations (the transitions between states are where the highest power draws occur) so I would speculate that you are using the same amount of power overall, unless you take into account the need for higher heat dissapation at higher transitions.
Can you point me to some references, or give me an answer on my previous questions.
You may be interested to know that some students at the University of Manitoba made a car that ran on grass in the late nineties, so this isn't a new idea. They won a few awards for it too, if I'm not mistaken.
Actually, I think it would be nice for them to have a strong female lead. Or I did before the whole Tomb Raider thing. But a niece or daughter would be good. It would give Indiana Jones a heart attack to see his daughter meeting men like him!
If exempt employees include engineers, most software engineers are also exempt. This is because the term engineer, as a professional, only applies to those who have taken an accredited engineering degree (or appropriate testing) and are legaly designated professional engineers.
Most comp-sci majors, regardless of their job titles, are not Engineers.
Could be wrong, but in this line you're saying
4000=40*10^3
Think that should be 4*10^3.
In your later post you say that means 4GB per song. Since the factor's off by 10, I guess that means it's really 400MB per song.
But then again... 40 * 10^15 (40PB) = 4 * 10 ^ 16 bytes
then becomes (40 * 10^16)bytes
So I guess that all evens out back to 4GB
But then again, I'm a geer, so I can't do simple math without a calculator and could be way off!
I think what many people want to know, is what finding, or not finding extraterrestrial life would mean to themselves, and to earth. Honestly, I don't think we can know ahead of time. I don't think we can prepare for the effects it will have. I don't think I will be around for our first contact should it ever arise, though I would love to see it happen, and think it would be interesting to see how humanity reacts.
What new questions and what new answers will this bring?
Then again, it's probably just as simple as 42.
I'm not bashing your point, more like a little fyi, but authors see royalties from libraries with photocopiers. At least they do in Canada. It's through an agreement called CanCopy, which legalises the copying of material - to an extent I believe. I don't think it's for the entire book.
Efficiency of a parallel computer considered to be
E=Ts/(n*Tp)
where Ts is the time to perform the computations serially, Tp is the the total time to perform the computations on the parallel machine and n is the number of parallel processing units.
It wouldn't take much to get a drastic improvement in efficiency simply by improving the time slightly for each parallel processer, especially for 1100 nodes.
I don't know how the benchmark program runs, but improving the communication time would imrove the efficiency as well.
It shouldn't take much to boost this by a few million flops.
Slashdot Mirror
You know, I used to be on the upgrade treadmill. I've got to get a new computer in order to play games etc. I've done two things that have changed the way I look at computers. First, I bought a Mac (Why? Because it is a great laptop for a engineer/programmer/student, and generally anyone. Besides the point anyway) I kept my PC, but it quickly stopped being used. Which brings me to my second point. I stopped playing PC games. Generally I just didn't have time, and then I didn't want to. I have a couple consoles that I use on rare occasions. Very rare occasions.
Now that I don't play PC games, there's no reason for me to own a PC. My powerbook does everything and more.
Anyone can use whatever they want, and I don't care, but I won't be going back the way of MS unless there's something that will make me want to switch. That doesn't seem to be happening yet.
Just a couple questions:
What is the power draw at 100% load on a single computer as compared to the same computing power on multiple computers at lower loads?
In other words, how many CPUs are needed at 10% load to achieve the same computing power as a single CPU running at 100% load, and how does the power draw compare?
Now, if you were to run the same computation on one computer at lower load levels (say let something run overnight if available) how does the power draw compare?
I have a hypothosis: Power draw increases with the number of operations (the transitions between states are where the highest power draws occur) so I would speculate that you are using the same amount of power overall, unless you take into account the need for higher heat dissapation at higher transitions.
Can you point me to some references, or give me an answer on my previous questions.
Thanks
You may be interested to know that some students at the University of Manitoba made a car that ran on grass in the late nineties, so this isn't a new idea. They won a few awards for it too, if I'm not mistaken.
Actually, I think it would be nice for them to have a strong female lead. Or I did before the whole Tomb Raider thing. But a niece or daughter would be good. It would give Indiana Jones a heart attack to see his daughter meeting men like him!
If exempt employees include engineers, most software engineers are also exempt. This is because the term engineer, as a professional, only applies to those who have taken an accredited engineering degree (or appropriate testing) and are legaly designated professional engineers.
Most comp-sci majors, regardless of their job titles, are not Engineers.
1MB = 10^6 bytes 4000MB = 40 * 10^6 * 10^ 3 = 4 * 10 ^ 10
Could be wrong, but in this line you're saying 4000=40*10^3
Think that should be 4*10^3.
In your later post you say that means 4GB per song. Since the factor's off by 10, I guess that means it's really 400MB per song.
But then again...
40 * 10^15 (40PB) = 4 * 10 ^ 16 bytes then becomes
(40 * 10^16)bytes
So I guess that all evens out back to 4GB
But then again, I'm a geer, so I can't do simple math without a calculator and could be way off!
I think what many people want to know, is what finding, or not finding extraterrestrial life would mean to themselves, and to earth. Honestly, I don't think we can know ahead of time. I don't think we can prepare for the effects it will have. I don't think I will be around for our first contact should it ever arise, though I would love to see it happen, and think it would be interesting to see how humanity reacts. What new questions and what new answers will this bring? Then again, it's probably just as simple as 42.
I'm not bashing your point, more like a little fyi, but authors see royalties from libraries with photocopiers. At least they do in Canada. It's through an agreement called CanCopy, which legalises the copying of material - to an extent I believe. I don't think it's for the entire book.
Efficiency of a parallel computer considered to be
E=Ts/(n*Tp)
where Ts is the time to perform the computations serially, Tp is the the total time to perform the computations on the parallel machine and n is the number of parallel processing units.
It wouldn't take much to get a drastic improvement in efficiency simply by improving the time slightly for each parallel processer, especially for 1100 nodes.
I don't know how the benchmark program runs, but improving the communication time would imrove the efficiency as well.
It shouldn't take much to boost this by a few million flops.