ah, but these "high school students" become customers if the roose is successful. each one paying a few grand (i haven't looked at tech's tuition, but colleges these days can cost tens of thousands per year, maybe more). if the amount of new revenue from an increased amount of students doesn't cover the cost, then other factors may. with the influx, they grow an increased chance of getting brilliant students (ontop of those they already have). these brilliants invent the next best thing, and say "it's all thanks to tech, and that wonderous supercomputer which convinced me to school there." press hears about this and writes up stories. then even more students read these "stories" and say "hey, i'd like to throw va tech a hundred grand or so to allow me to school near this supercomputer." and thus, the circle grows into a spiral. and the spiral swirls all the way to the bank.
No. Just no.
Sure having a super computer puts a nice spit and polish on Va Tech but no chairman of a department spends 5 million on a super computer to attract undergraduates.
Well the main reason is 3D porn...it gets pretty hard to realistic 3d porn when it's more then one girl! I mean sure missle trajectory and medical crap, but if you want to know the real reason reread my first sentence
This has gotta help convert students to be "Mac for life" as well. I mean think about it, using a supercomputer to get your research done? Youre gonna want what you had that was that fast when you were in school. This is nothing but a great big win for Apple now and into the future.
Back when I was a student there my roommate was required to buy a Mac for CS. Then my other roommate who entered CS later had to buy an amiga
I don't know what they use today buy I'd guess Virusdows or Linux. What self respecting CS student would have an x86 box and not have a Linux install on it?
as i said b4 i don't know much about the supercomputers but i know thats not why
You know nothing about how universities work then. You better believe improving possibilities for grant funding was a MAJOR reason for this project. It may have been the primary reason. That's how important grant money is.
That and they have the opportunity to rent out the CPU time for big big bucks.
You know nothing about how universities work then. You better believe improving possibilities for grant funding was a MAJOR reason for this project. It may have been the primary reason. That's how important grant money is.
That and they have the opportunity to rent out the CPU time for big big bucks.
no way... that would be a HUGE waste of money and time...
apple would NOT give them the first batch of dual 2.0s only to take them back later...
and there wouldn't be much of a performance increase by going to the Xserve... besides they just want cycles... not too much concerned about reliability... the Xserve would be overkill for them...
sure it would save a lot of space, but they already set it up...
well, it seems that I was wrong....
but I still think it is a waste of time...
does anyone have a link to the new deal's details?
I dont think they realize you are 13, although it is clear from your profile/naivite.
As Brad said there are a lot of computations that simply cannot be done on your average desktop, and I will dare say there will always be simply because brute force calculations (those calculations where the programmer doesnt think and tries every possible iteration) are possibly always easier to program. VT's intention is the same as a lot of other large perhaps second tier schools, focusing on one specific subject matter (in this case supercomputing/clustering) can bring in grants and private money that would ordinarily not be readily available for an institution that does not have a major singular focus. Selling computation time would probably not regenerate the cost over the long run, since these machines will need to be replaced very frequently to stay competitive. The operation cost is most likely all they expect from selling on time. It is a research project after all and not a buisness. BTW, most universities/colleges are not for profit institutions so tuition shouldnt really be considered a source of money.
I dont think they realize you are 13, although it is clear from your profile/naivite.
As Brad said there are a lot of computations that simply cannot be done on your average desktop, and I will dare say there will always be simply because brute force calculations (those calculations where the programmer doesnt think and tries every possible iteration) are possibly always easier to program. VT's intention is the same as a lot of other large perhaps second tier schools, focusing on one specific subject matter (in this case supercomputing/clustering) can bring in grants and private money that would ordinarily not be readily available for an institution that does not have a major singular focus. Selling computation time would probably not regenerate the cost over the long run, since these machines will need to be replaced very frequently to stay competitive. The operation cost is most likely all they expect from selling on time. It is a research project after all and not a buisness. BTW, most universities/colleges are not for profit institutions so tuition shouldnt really be considered a source of money.
thanks for the clearer explanation i think i get it now
first of all, isn't this what the Xserve is for? clusters!!
Va Tech just used desktops cuz they had a deadline (for a Grant, they probably got, so that's what the original investment was for) and the Xserve G5 wasn't ready! They can fit more Xserves than desktops so they can expand, nice...
i'm sure this upgrade is well thought out (after all, they made the 3rd fastest supercomputer in the first place!) and this is will be the real model of supercomputer that folks will be copying...
worrying about selling desktops and such is silly...Va Tech is making enough money w/ contracts to outside government organizations that are buying time w/ the supercomputer. Also, they are writing the book on cluster supercomputers, so it's nice to see they are willing to accept the challenge , not be lazy and complacent...
There's no need to have two active threads about VT's Xserves.
<pet peeve>
It's Xserve, folks, not X-Serve or XServe or xServe. I would have thought that you guys (not intentionally pointing fingers) could get that straight when so many people here are the ones to correct others for saying MAC or I-Mac or iMAC, etc.
</pet peeve>
So, I'm merging the two threads and correcting the spelling on the title.
Comments
Originally posted by thuh Freak
ah, but these "high school students" become customers if the roose is successful. each one paying a few grand (i haven't looked at tech's tuition, but colleges these days can cost tens of thousands per year, maybe more). if the amount of new revenue from an increased amount of students doesn't cover the cost, then other factors may. with the influx, they grow an increased chance of getting brilliant students (ontop of those they already have). these brilliants invent the next best thing, and say "it's all thanks to tech, and that wonderous supercomputer which convinced me to school there." press hears about this and writes up stories. then even more students read these "stories" and say "hey, i'd like to throw va tech a hundred grand or so to allow me to school near this supercomputer." and thus, the circle grows into a spiral. and the spiral swirls all the way to the bank.
No. Just no.
Sure having a super computer puts a nice spit and polish on Va Tech but no chairman of a department spends 5 million on a super computer to attract undergraduates.
Originally posted by PS5533
no seriousley...what is it for?
Well the main reason is 3D porn...it gets pretty hard to realistic 3d porn when it's more then one girl! I mean sure missle trajectory and medical crap, but if you want to know the real reason reread my first sentence
I don't know what they use today buy I'd guess Virusdows or Linux. What self respecting CS student would have an x86 box and not have a Linux install on it?
Originally posted by PS5533
as i said b4 i don't know much about the supercomputers but i know thats not why
You know nothing about how universities work then. You better believe improving possibilities for grant funding was a MAJOR reason for this project. It may have been the primary reason. That's how important grant money is.
That and they have the opportunity to rent out the CPU time for big big bucks.
Oh yea, and all the other stuff you guys said too. 8)
Originally posted by torifile
You know nothing about how universities work then. You better believe improving possibilities for grant funding was a MAJOR reason for this project. It may have been the primary reason. That's how important grant money is.
That and they have the opportunity to rent out the CPU time for big big bucks.
geez sorry
Originally posted by DMBand0026
Is anyone else thinking that this could be a big help in the protien folding thing that this thread is talking about? That would be sweet.
Oh yea, and all the other stuff you guys said too. 8)
I wonder how much could be done in just a day!
Originally posted by Paul
no way... that would be a HUGE waste of money and time...
apple would NOT give them the first batch of dual 2.0s only to take them back later...
and there wouldn't be much of a performance increase by going to the Xserve... besides they just want cycles... not too much concerned about reliability... the Xserve would be overkill for them...
sure it would save a lot of space, but they already set it up...
well, it seems that I was wrong....
but I still think it is a waste of time...
does anyone have a link to the new deal's details?
Originally posted by PS5533
geez sorry
I dont think they realize you are 13, although it is clear from your profile/naivite.
As Brad said there are a lot of computations that simply cannot be done on your average desktop, and I will dare say there will always be simply because brute force calculations (those calculations where the programmer doesnt think and tries every possible iteration) are possibly always easier to program. VT's intention is the same as a lot of other large perhaps second tier schools, focusing on one specific subject matter (in this case supercomputing/clustering) can bring in grants and private money that would ordinarily not be readily available for an institution that does not have a major singular focus. Selling computation time would probably not regenerate the cost over the long run, since these machines will need to be replaced very frequently to stay competitive. The operation cost is most likely all they expect from selling on time. It is a research project after all and not a buisness. BTW, most universities/colleges are not for profit institutions so tuition shouldnt really be considered a source of money.
Originally posted by DMBand0026
Is anyone else thinking that this could be a big help in the protien folding thing that this thread is talking about? That would be sweet.
Oh yea, and all the other stuff you guys said too. 8)
Um, these computers (meaning supercomputers in general) are being used to 'solve' the protein folding problem already.
Originally posted by billybobsky
I dont think they realize you are 13, although it is clear from your profile/naivite.
As Brad said there are a lot of computations that simply cannot be done on your average desktop, and I will dare say there will always be simply because brute force calculations (those calculations where the programmer doesnt think and tries every possible iteration) are possibly always easier to program. VT's intention is the same as a lot of other large perhaps second tier schools, focusing on one specific subject matter (in this case supercomputing/clustering) can bring in grants and private money that would ordinarily not be readily available for an institution that does not have a major singular focus. Selling computation time would probably not regenerate the cost over the long run, since these machines will need to be replaced very frequently to stay competitive. The operation cost is most likely all they expect from selling on time. It is a research project after all and not a buisness. BTW, most universities/colleges are not for profit institutions so tuition shouldnt really be considered a source of money.
thanks for the clearer explanation i think i get it now
Va Tech just used desktops cuz they had a deadline (for a Grant, they probably got, so that's what the original investment was for) and the Xserve G5 wasn't ready! They can fit more Xserves than desktops so they can expand, nice...
i'm sure this upgrade is well thought out (after all, they made the 3rd fastest supercomputer in the first place!) and this is will be the real model of supercomputer that folks will be copying...
worrying about selling desktops and such is silly...Va Tech is making enough money w/ contracts to outside government organizations that are buying time w/ the supercomputer. Also, they are writing the book on cluster supercomputers, so it's nice to see they are willing to accept the challenge , not be lazy and complacent...
<pet peeve>
It's Xserve, folks, not X-Serve or XServe or xServe. I would have thought that you guys (not intentionally pointing fingers) could get that straight when so many people here are the ones to correct others for saying MAC or I-Mac or iMAC, etc.
</pet peeve>
So, I'm merging the two threads and correcting the spelling on the title.
System X Upgrade In Progress: