The idea that this is just a marketing ploy by Sony has been beaten to death several times by now.
But there is one thing to keep in mind: Less than a decade ago you had to sign an agreement that you were not from an "unfriendly" country like North Korea, PRC,... when you wanted access to a machine which had approximately the same floating point performance as the PS2 now.
Now just build a nice machine with about 200 EEs and voila 1.2 TeraFlops peak performance (which means that you are guaranteed not to reach or exceed the performance). Now get say 1024 of those babies and 1 TeraFlop in the linpack benchmark should be within reach (currently you would make the top 10 of the Top500 list of supercomputers see: http://www.top500.org).
Reasoning along those lines it makes perfectly sense to restrict export.
Whether it makes sense to restrict export of a commodity item like the PS2 is a completely different story.
P.S.: Does anyone know whether the 6 GFlops are single or double precision - in the latter case a 1024 EE machine might be a real nice toy;-) P.P.S.: Linpack is double precision - but do those goverment burocrats do know the difference?
UMTS the 3rd generation cellular system will deliver bandwiths up to 2Mbps and is a successor of today GSM systems. In the case of Iridium the evolution of mobile phones rendered this business obosolete. I wonder now what is going to happen to iSKY. Especially since the first UMTS operations will be launched in 2002. UMTS supposedly will be comprised of a mixture of terrestrial and (optionally) satellite based relays. Finally it will be (hopefully) the first world wide standard for mobile communications. I wish them good luck, but I am afraid that this might be just another Iridium.
For more information on UMTS, have a look at: http://www.cellular.co.za/
As a follow up: In principle proofing a system against such things is quite simple: just limit the number of processes on a per user basis (with AIX for example this is possible). If this seems to be not an option - then reserve a few processes for root use only. These things ought to be configurable...
So maybe that specific system I mentioned in my first post was just poorly configured (however the guys responsible for setting it up in the first place probably just installed it "out of the box") - so my critcism is still valid.
Besides I am too old to get into that OS-war thing anymore - I couldn't care less if a box were running Unix (SysV, BSD, OSF1, anything "compatible" like AIX or Linux), VMS, MVS or (cough!) NT. As long as it delivers it is OK - if it gets in the way of doing business, well... Picture about 200 people thinking that their bonus is at risk just because of a bloody computer fault and all of them think that you are responsible!
Any script like this (you might as well use csh) will do the job: File: fork.ksh -------------------------------------------- #!/bin/ksh
while [ 1 ]; do echo "another loop" nohup fork.ksh & done -------------------------------------------- Let this baby run for a while and then try a graceful shutdown/logoff from their UI...
In 1989 when I got access to a Unix box the first time in my life (it was a SUN running SunOS - another BSD) I made a mistake in my homework assignment and crashed the departments SUN used for the course. Ever since, one of the first "tests" when accessing a new box is this little script...
[and now for the flame:-( ] If you would have been able to control your temper, you would have noticed that I wrote the word "deliberately". Any machine or software can be crashed deliberately (keyword here is: boundary conditions) - I was simple annoyed that it was still _that_ easy (given the nimbus Apple is sporting). And if such a machine is to be used in an investment banks trading room, competing against a NT-workstation running the same software - better make sure it is really stable!
... disregarding the fact that (unless Apple got their act together in the last 6 month) it gives BSD a bad name.
We had one of those boxes (with OS-X Server) for testing a couple of months ago. It took me about 10 minutes to crash that machine: no root access necessary, just plain old shell scripting (details upon request).
Too bad Apple isn't that keen on real aggressive pricing, with increased stability this could lead to 'NIXes finally toppling the MS dominance...
I hate to say that, but Cray had been in trouble for more than a decade. Already in the early '90s Cray had problems keeping up with the vector processors from the big 3 japanese companies (NEC, Fujitsu and Hitachi) - at that time their only advantage was that they had a lead in their software tools - raw processing power wise the NEC SX3 (I am not even talking about the Monte4 variant) ran circles around the Y-MPs then. So whenever someone (in the US mainly) dared to order a vector machine from someone else than Cray, they would start whining about "dumping" and unfair competition. The other problem which emerged for Cray was the advent of high performace workstations. In some cases these superscalar workstations could outperform any vector processor (not all algorithms vectorize). So their machines also were attacked from the low end. To make things worse the advent of MPP systems (sometimes using these aforementioned processors) put additonal pressure from the high end on Cray. But worst of all: the cold war ended and the majority of supercomputer customers did not have that much money to spend anymore. The remaining money went into the real hot (at that time) MPP machines (TMC's CM2 and CM5, Intel's Hypercube and Paragon,...). But even TMC went out of business (although I still think the CM2 was one of the coolest machines at that time;-) ). So finally (having lost the race to build the fastest vector processor) Cray came up with the T3D (the E came later). But using off the shelf CPU's doesn't make these machines any cheaper - now the communication is much more complex, the software (just to run those babies) is something entirely new and finally: not all problems can be parallelized! IMHO nowadays supercomputers can only be build by the big players in the market, who could afford to loose money on that: write off some to marketing and some to R&D. Cray would have been better off, if they were bought by some of the big players. I wish Tera/Cray good luck - but I am afraid it will be just another life extension for one of the most influential computer manufacturers of the past. Without serious amounts of money this will just defer the inevitable... Just a note: supercomputing had been in trouble since the late 80s - anyone remember the ETA10?
I agree that a cluster does not make a supercomputer - but claiming that a supercomputer has to use shared memory is something different. A supercomputer can range from a single processor vector machine, a shared memory machine with a moderate amount of processors (e.g. the gold old Y-MP) to a MPP shared nothing architecture (optionally with vector units). The difference between clusters and MPP shared nothing architectures is the underlying communication. Clusters usually utilize off the shelf equipment while those MPP beasts use a specialized communications architecture. (keywords: meshes, hypercubes, fat trees, full crossbars (if you have the money) etc.). So please don't confuse multiprocessor shared memory machines with supercomputers.
This station had been deployed in 1986, it has been specced for about 5 years of operation, so now it is at about 300% of its projected lifetime. Any other operation like this (those two kosmonauts IMHO are risking their lives) would be called irresponsible... Refurbishing MIR to extend its lifetime again will be costly... The russian modules for the new space station allegedly (that is what the "news" feed me) not meeting spec and delayed due to funding problems - to me this whole MIR operation looks like a publicity stunt. IMHO the money would be better spent on the new station...
Re:Selling philosophy disguised as science...
on
The Mind of God
·
· Score: 1
Point well taken. I am not discounting philosophy per se, but I am not very happy if things get mixed up. (natural) sciences and philosophy are "two different beasts" - IMHO they are asking different questions. When this distinction gets fuzzy, people from both sides of the fence (usually) start producing less than stellar "results".
Selling philosophy disguised as science...
on
The Mind of God
·
· Score: 3
Davies is a decent pop-science writer and probably a decent scientist too - but mixing up philosophy (religion is just a subset) and physics doesn't make sense. The difference between science and philosophy is, that physical models can be empirically falsified, philosophical musings cannot (per definition). Occam's razor still applies: "If you can't explain it, don't explain it by means you still cannot explain..."
Slashdot Mirror
The idea that this is just a marketing ploy by Sony has been beaten to death several times by now.
... when you wanted access to a machine which had approximately the same floating point performance as the PS2 now.
;-)
But there is one thing to keep in mind: Less than a decade ago you had to sign an agreement that you were not from an "unfriendly" country like North Korea, PRC,
Now just build a nice machine with about 200 EEs and voila 1.2 TeraFlops peak performance (which means that you are guaranteed not to reach or exceed the performance). Now get say 1024 of those babies and 1 TeraFlop in the linpack benchmark should be within reach (currently you would make the top 10 of the Top500 list of supercomputers see: http://www.top500.org).
Reasoning along those lines it makes perfectly sense to restrict export.
Whether it makes sense to restrict export of a commodity item like the PS2 is a completely different story.
P.S.: Does anyone know whether the 6 GFlops are single or double precision - in the latter case a 1024 EE machine might be a real nice toy
P.P.S.: Linpack is double precision - but do those goverment burocrats do know the difference?
UMTS the 3rd generation cellular system will deliver bandwiths up to 2Mbps and is a successor of today GSM systems.
In the case of Iridium the evolution of mobile phones rendered this business obosolete. I wonder now what is going to happen to iSKY. Especially since the first UMTS operations will be launched in 2002. UMTS supposedly will be comprised of a mixture of terrestrial and (optionally) satellite based relays. Finally it will be (hopefully) the first world wide standard for mobile communications.
I wish them good luck, but I am afraid that this might be just another Iridium.
For more information on UMTS, have a look at:
http://www.cellular.co.za/
(Note: I am not affiliated with them in any way.)
Thanks for the support WiseWeasel ;-)
As a follow up:
In principle proofing a system against such things is quite simple: just limit the number of processes on a per user basis (with AIX for example this is possible). If this seems to be not an option - then reserve a few processes for root use only.
These things ought to be configurable...
So maybe that specific system I mentioned in my first post was just poorly configured (however the guys responsible for setting it up in the first place probably just installed it "out of the box") - so my critcism is still valid.
Besides I am too old to get into that OS-war thing anymore - I couldn't care less if a box were running Unix (SysV, BSD, OSF1, anything "compatible" like AIX or Linux), VMS, MVS or (cough!) NT. As long as it delivers it is OK - if it gets in the way of doing business, well... Picture about 200 people thinking that their bonus is at risk just because of a bloody computer fault and all of them think that you are responsible!
Any script like this (you might as well use csh) will do the job:
:-( ]
File: fork.ksh
--------------------------------------------
#!/bin/ksh
while [ 1 ]; do
echo "another loop"
nohup fork.ksh &
done
--------------------------------------------
Let this baby run for a while and then try a
graceful shutdown/logoff from their UI...
In 1989 when I got access to a Unix box the first time in my life (it was a SUN running SunOS - another BSD) I made a mistake in my homework assignment and crashed the departments SUN used for the course. Ever since, one of the first "tests" when accessing a new box is this little script...
[and now for the flame
If you would have been able to control your temper, you would have noticed that I wrote the word "deliberately". Any machine or software can be crashed deliberately (keyword here is: boundary conditions) - I was simple annoyed that it was still _that_ easy (given the nimbus Apple is sporting).
And if such a machine is to be used in an investment banks trading room, competing against a NT-workstation running the same software - better make sure it is really stable!
So now the main question: Who is the dweeb here?
... disregarding the fact that (unless Apple got their act together in the last 6 month) it gives BSD a bad name.
We had one of those boxes (with OS-X Server) for testing a couple of months ago. It took me about 10 minutes to crash that machine: no root access necessary, just plain old shell scripting (details upon request).
Too bad Apple isn't that keen on real aggressive pricing, with increased stability this could lead to 'NIXes finally toppling the MS dominance...
I hate to say that, but Cray had been in trouble for more than a decade. Already in the early '90s Cray had problems keeping up with the vector processors from the big 3 japanese companies (NEC, Fujitsu and Hitachi) - at that time their only advantage was that they had a lead in their software tools - raw processing power wise the NEC SX3 (I am not even talking about the Monte4 variant) ran circles around the Y-MPs then. So whenever someone (in the US mainly) dared to order a vector machine from someone else than Cray, they would start whining about "dumping" and unfair competition. The other problem which emerged for Cray was the advent of high performace workstations. In some cases these superscalar workstations could outperform any vector processor (not all algorithms vectorize). So their machines also were attacked from the low end. To make things worse the advent of MPP systems (sometimes using these aforementioned processors) put additonal pressure from the high end on Cray. But worst of all: the cold war ended and the majority of supercomputer customers did not have that much money to spend anymore. The remaining money went into the real hot (at that time) MPP machines (TMC's CM2 and CM5, Intel's Hypercube and Paragon, ...). But even TMC went out of business (although I still think the CM2 was one of the coolest machines at that time ;-) ). So finally (having lost the race to build the fastest vector processor) Cray came up with the T3D (the E came later). But using off the shelf CPU's doesn't make these machines any cheaper - now the communication is much more complex, the software (just to run those babies) is something entirely new and finally: not all problems can be parallelized! IMHO nowadays supercomputers can only be build by the big players in the market, who could afford to loose money on that: write off some to marketing and some to R&D. Cray would have been better off, if they were bought by some of the big players. I wish Tera/Cray good luck - but I am afraid it will be just another life extension for one of the most influential computer manufacturers of the past. Without serious amounts of money this will just defer the inevitable... Just a note: supercomputing had been in trouble since the late 80s - anyone remember the ETA10?
I agree that a cluster does not make a supercomputer - but claiming that a supercomputer has to use shared memory is something different. A supercomputer can range from a single processor vector machine, a shared memory machine with a moderate amount of processors (e.g. the gold old Y-MP) to a MPP shared nothing architecture (optionally with vector units). The difference between clusters and MPP shared nothing architectures is the underlying communication. Clusters usually utilize off the shelf equipment while those MPP beasts use a specialized communications architecture. (keywords: meshes, hypercubes, fat trees, full crossbars (if you have the money) etc.). So please don't confuse multiprocessor shared memory machines with supercomputers.
This station had been deployed in 1986, it has been specced for about 5 years of operation, so now it is at about 300% of its projected lifetime. Any other operation like this (those two kosmonauts IMHO are risking their lives) would be called irresponsible... Refurbishing MIR to extend its lifetime again will be costly... The russian modules for the new space station allegedly (that is what the "news" feed me) not meeting spec and delayed due to funding problems - to me this whole MIR operation looks like a publicity stunt. IMHO the money would be better spent on the new station...
Point well taken. I am not discounting philosophy per se, but I am not very happy if things get mixed up. (natural) sciences and philosophy are "two different beasts" - IMHO they are asking different questions. When this distinction gets fuzzy, people from both sides of the fence (usually) start producing less than stellar "results".
Davies is a decent pop-science writer and probably a decent scientist too - but mixing up philosophy (religion is just a subset) and physics doesn't make sense. The difference between science and philosophy is, that physical models can be empirically falsified, philosophical musings cannot (per definition). Occam's razor still applies: "If you can't explain it, don't explain it by means you still cannot explain..."