Land line phone wire taps are not administered by the FBI but by the local phone company. Yes, conceivably cell phone taps could be done without cooperation of the phone company, but only at great expense of CPU time to attempt to break the security on those phones. I'm not saying it can't be done, but the effort required is non-trivial, and thus if more than a few such wiretaps are desired, it becomes a tremendous burden to break each and every cryptographic key protecting the voice convos.
Comments about open source security holes bring to mind the Fuzz tests of 1990 and 1995. (Yes, I know, somewhat dated, hopefully it will be updated). In the 1995 fuzz test, a Linux distribution was tested among other systems. (I don't believe a *BSD distro was tested). Interestingly enough, the Open source stuff tested performed far better against the Fuzz test than closed source did (with NeXT performing the worst by far.) If Open Source encouraged security holes, one would expect that at a minimum, fuzz tests on open source software would produce a comparable failure rate to non Open Source UN*X systems tested (including Solaris and HP-UX among others).
We can talk all we want about logical reasons why open source software does not produce greater security holes, but empirical comparison studies of a stable distribution are in many ways far more convincing.
For more information on the Fuzz test, see
http://www.cs.wisc.edu/~bart/fuzz/fuzz.html
(Note especially the conclusions of the 2000 NT Fuzz test)
I really shouldn't give this comment a reply, but there are a few gross misconceptions that were not addressed by others here.
Mutations in and of themselves are not necessarily bad. Self organizational structure is hardly uncommon. (Just create a non-linear feedback loop, let it run a few generations, and unless you were careful to try for pseudo-random nature, you'll probably get structure appearing. Evolution is not the creation of life from non-life. Intelligent design proponents invariably get caught in the cycle of the origin of the designer. Once it is accepted that said designer was not designed but arose spontaneously, the need for a designer to give rise to complex organic systems disappears. (That's the sound bite sized reason why the argument from design is not very good.) YOur argument seems to be based on the highly flawed concept that self organizational structure is somehow rare or unusual.
As someone who argues extensively with those who "have doubts about evolution", I realize that almost every objection is based on argument from incredulity based on misunderstanding of what the term evolution even means, also what the word theory means. There are a small number who actually have objections based not on a misunderstanding of what evolutionary science is, but they are so rare in my experience as to be almost negligable.
Oh, as evidence for non-harmful mutations, I have a relative who when a permanent tooth was pulled, another one came in place, another relative with a fully functional extra half kidney (Could have full failure of one kidney and _still_ have suitable kidney function to not need dialysis).
As I recall from some other work I've done, one application that scales well to MPP is heat distribution of a component that generates heat and has a custom heat sink. Once you can model nuclear explosions, it doesn't take much imagination to come up with other molecular modelling problems that can be used for such a system. I'm sure someone could write a model of a biological system that had little enough communication between regions of the biological system as to be feasible in a MPP design. (The uses of being able to model arbitrary and theoretical biological regions should be obvious). Then there are the classic mathematical problems that could find a MPP supercomputer useful. I'm not sure offhand how I'd parallelize the code, but the twin prime conjecture would be one example. (Probably parallelize regions of the number plane). Research into the factoring of large numbers is also feasible (of course we don't know any practical application for that...)
Security is one good reason for having it on a single computer. I somehow suspect that the people who would like to model nuclear explosions wouldn't like their models and information available to various other parties. Keeping it on a dedicated beowulf cluster helps contain the information in a way that even doing distributed processing across a classified network cannot do. As other posters also pointed out, these processes require more interprocess communication than is convienient in the distributed processing model, but not so much that vector processing or the interconnections in say the Cray T-3E would require. (And at a fraction of the cost of a T-3, I suspect many people are more forgiving of speed issues.)
Depends on the job. Just because a problem is thrown at a supercomputer doesn't mean it necessarily requires massive interprocesser communication. It was the number of jobs appearing that could be readily broken down into multiple jobs that gave rise to the Thinking Machines line of MPP supercomputers. Nuclear explosions are surprisingly independent in terms of level of interprocessor communication.
Obviously you can't just throw 500 PCs running the Beowulf kernel and call it a supercomputer. You do need dedicated high speed networking, and clearly not all jobs parallelize readily to that model. For a special purpose computer however, Beowulf is quite acceptable in the case of relatively minimal communication between the processes compared to the communication needed in say a vector processor based supercomputer.
One company I've heard of that builds these Beowulf clusters is http://www.hpti.com/ From what I've heard, they do use some heavy duty connections between the nodes.
Slashdot Mirror
Land line phone wire taps are not administered by the FBI but by the local phone company. Yes, conceivably cell phone taps could be done without cooperation of the phone company, but only at great expense of CPU time to attempt to break the security on those phones. I'm not saying it can't be done, but the effort required is non-trivial, and thus if more than a few such wiretaps are desired, it becomes a tremendous burden to break each and every cryptographic key protecting the voice convos.
Comments about open source security holes bring to mind the Fuzz tests of 1990 and 1995. (Yes, I know, somewhat dated, hopefully it will be updated). In the 1995 fuzz test, a Linux distribution was tested among other systems. (I don't believe a *BSD distro was tested). Interestingly enough, the Open source stuff tested performed far better against the Fuzz test than closed source did (with NeXT performing the worst by far.) If Open Source encouraged security holes, one would expect that at a minimum, fuzz tests on open source software would produce a comparable failure rate to non Open Source UN*X systems tested (including Solaris and HP-UX among others).
We can talk all we want about logical reasons why open source software does not produce greater security holes, but empirical comparison studies of a stable distribution are in many ways far more convincing.
For more information on the Fuzz test, see
http://www.cs.wisc.edu/~bart/fuzz/fuzz.html
(Note especially the conclusions of the 2000 NT Fuzz test)
I really shouldn't give this comment a reply, but there are a few gross misconceptions that were not addressed by others here.
Mutations in and of themselves are not necessarily bad. Self organizational structure is hardly uncommon. (Just create a non-linear feedback loop, let it run a few generations, and unless you were careful to try for pseudo-random nature, you'll probably get structure appearing. Evolution is not the creation of life from non-life. Intelligent design proponents invariably get caught in the cycle of the origin of the designer. Once it is accepted that said designer was not designed but arose spontaneously, the need for a designer to give rise to complex organic systems disappears. (That's the sound bite sized reason why the argument from design is not very good.) YOur argument seems to be based on the highly flawed concept that self organizational structure is somehow rare or unusual.
As someone who argues extensively with those who "have doubts about evolution", I realize that almost every objection is based on argument from incredulity based on misunderstanding of what the term evolution even means, also what the word theory means. There are a small number who actually have objections based not on a misunderstanding of what evolutionary science is, but they are so rare in my experience as to be almost negligable.
Oh, as evidence for non-harmful mutations, I have a relative who when a permanent tooth was pulled, another one came in place, another relative with a fully functional extra half kidney (Could have full failure of one kidney and _still_ have suitable kidney function to not need dialysis).
As I recall from some other work I've done, one application that scales well to MPP is heat distribution of a component that generates heat and has a custom heat sink. Once you can model nuclear explosions, it doesn't take much imagination to come up with other molecular modelling problems that can be used for such a system. I'm sure someone could write a model of a biological system that had little enough communication between regions of the biological system as to be feasible in a MPP design. (The uses of being able to model arbitrary and theoretical biological regions should be obvious). Then there are the classic mathematical problems that could find a MPP supercomputer useful. I'm not sure offhand how I'd parallelize the code, but the twin prime conjecture would be one example. (Probably parallelize regions of the number plane). Research into the factoring of large numbers is also feasible (of course we don't know any practical application for that...)
Security is one good reason for having it on a single computer. I somehow suspect that the people who would like to model nuclear explosions wouldn't like their models and information available to various other parties. Keeping it on a dedicated beowulf cluster helps contain the information in a way that even doing distributed processing across a classified network cannot do. As other posters also pointed out, these processes require more interprocess communication than is convienient in the distributed processing model, but not so much that vector processing or the interconnections in say the Cray T-3E would require. (And at a fraction of the cost of a T-3, I suspect many people are more forgiving of speed issues.)
Depends on the job. Just because a problem is thrown at a supercomputer doesn't mean it necessarily requires massive interprocesser communication. It was the number of jobs appearing that could be readily broken down into multiple jobs that gave rise to the Thinking Machines line of MPP supercomputers. Nuclear explosions are surprisingly independent in terms of level of interprocessor communication.
Obviously you can't just throw 500 PCs running the Beowulf kernel and call it a supercomputer. You do need dedicated high speed networking, and clearly not all jobs parallelize readily to that model. For a special purpose computer however, Beowulf is quite acceptable in the case of relatively minimal communication between the processes compared to the communication needed in say a vector processor based supercomputer.
One company I've heard of that builds these Beowulf clusters is http://www.hpti.com/ From what I've heard, they do use some heavy duty connections between the nodes.