If you read the article, McAffe assumes that if you do not get 100% right you are
at risk (only 3% got it all right). I took the test and got one right. I don't
stray into these kinds of sites, and I use Linux, and I don't browse websites as root, and
I don't download "free" stuff from the Internet (except rpms, pdfs, and the like from trusted sites, yes I know I'm boring)
Like most, I have no idea what a dangerous site
looks like and I assume most people, like me, cannot tell the difference so
if it is a fair test, then the majority should have had 4 right. (they were
picking randomly). Disclosure of the full results should show this, otherwise it
is bogas test.
What does this mean? Well, apart from the misuse of statistics for marketing purposes,
I think is more of the "blaming (charging) the victim" marketing mentality. For
the average users, they have been told "Here is this great thing called the Internet,
go forth and enjoy the new world". Then they are told, "You are so stupid you can't tell
which site will wreak your computer, you need to buy our software."
The problem needs to placed on software allows such unsanctioned downloads (i.e. the browser and OS). Of course you cannot stop the user from clicking where they should not, but
with the current majority of systems running Swiss cheese software like IE users will
continue to get infected.
There is no way the average user can tell if a well designed website is indeed a mal-site. Until users start demanding accountability for poorly designed software, it will continue.
Err.. I said "(for almost all applications) ". Please provide some applications
that use this method. I'm not trying to troll, I have a genuine interest in
heterogeneous clusters -- and how to use them efficiently for applications
that currently run best on homogeneous systems.
Before everyone gets all "top500" over this idea, here is little
cold water on the realities of cluster computing. Yes large number sof CPUs are great, but the software has to be there
to use them. Plus, for anything other than rendering, a heterogeneous cluster is limited by its slowest node (for almost all applications)
It is the cluster equivalent of how many people you can stuff into phone booth. If you have nothing better to do, well why not I guess.
Cray finally figured it out. I have been saying for years:
HPC/Beowulf clusters are about building machines around problems
That is why Clusters are such a powerful paradigm. If your problem needs more
processors/memory/bandwidth/data access, you can design a cluster to
fit your problem and only buy what your need. In the past you had to buy
a large supercomputer with lots of engineering you did not need. Designing clusters
is an art, but the payoff is very good price-to-performance. I even
wrote an article on
Cluster Urban Legends the explains many of these issues.
The transputer was not multi-core. It could be connected to other transputers very easily, however. What it did have that made it rather interesting was process management in hardware.
Non-computer people (and computer people) should understand that computers are machines. Used properly they should save time and money and augment the user. If a machine is wasting time because it breaks too often, is too hard to learn and use, easily jeopardizes your security and personal information, and just becomes a hassle, then complain loudly. Ask for you money back and remember just because it seems like a "complicated" device that does magical things, that is no excuse for it not to work as promised. Most people don't know how a car works, but would never tolerate cars that worked liked computers.
If everyone has that kind of attitude, computers would be different. However, computers today are a "time sink" basically running broken software. Unless you are a specialist with adequate training, computers are a waste of most peoples time. Use a pencil and paper, calculator, and typewriter, it is more efficient. Sounds silly? How many times have you had to tell someone how do use a word processor? And how many copies did they print (waiting paper), before they got that hanging indent right (using spaces) only to find that when they change the font and the document falls apart. Need I continue...
The big void is going to be in parallel and distributed computing. By 1998 desktops will
have 4 core processors, workstations with two sockets will have 8 cores.
Beyond multitasking, existing software cannot use these processors.
Not that
every program needs to use extra CPUs, but developers who have experienced
continued speed "free lunch" improvements are going to hit a wall unless they start thinking
in terms of threads, OpenMP, and MPI. You can check out Cluster Monkey for infromation on cluster computing which has been dealing with these issues for the last ten years.
This is not like telling students some new theory that someone thought up 5 minutes ago. People have believed in Christ for over 2000 years. It seems like it should be mentioned in the biology class.
I'm all for it. As long as they teach my religion as well. And by the way, although my religion is new to me, it has existed in since the beginning of time. It was created "new" only recently.
This is kind of odd. Burton Smith is not really a cluster guy, although he probably knows his way around HPC (High Performance Computing). Cray is not really a cluster company (except for
the system they bought from Octiga Bay deal). If you want to read a review of what Bill Gates said at the recent Supercomputing conference, check out
Where is the Cluster? at Cluster Monkey.
The writer is (deliberately) confusing open source with open standards. Open standards are how progress happens. Take for example ASTM Think what would happen a a large construction company owned and wrote the ASTM building codes. This is the problem with a proprietary document format from one company. Microsot is free to implement OpenDoc, but then is will reduce their "lock in".
Stop debating creationism vs Darwinism. I want equal time (and coroprate support) for the FSM Theory (and religion). I just spent the last night working on education material for my new found religion.
This is very interesting. The Cell has a very non-standard architecture, but it can be used in a very powerful way. The key is software and thus, an emulation SDK is really important for a new architecture. From and HPC (High Performance Computing) prospective, these chips could be very powerful.
The real question is whether the the PS3 will have an Linux hard disk option like the PS2. If that is the case, it may be the cheapest way to get actual development hardware.
There are some very good reasons why Linux is on Clusters. Having
been in the HPC business for over 17 years, I think the reasons
are not very obvious to many outside the field. To spread the word I wrote .
Why Linux On Clusters which
pretty much tells it like it is.
A good question. I believe at one point Gorden Bell said "anything with 6 zeros in the price".
The lines have blurred due to clusters. My definition is "a collection of hardware that provides a non-trivial level of performance on a single problem" Of course, "non trivial" has various interpretations. And, working toward solving a single problem is important. Rendering is
a trivial parallel application as it is really a bunch of small independent problems. Most supercomputer applications would probably run "sub-optimal" on this system (I assume it has GigE
as an interconnect)
because they require much more processor to processor communication. BTW, I run the ClusterMonkey site that talks about clusters and HPC if you want to learn more about clusters.
boosting its power by 50 per cent to create a supercomputer with the equivalent power of nearly 15,000 PCs
Such statements are utter nonsense. First, 15,000 PC's - what kind of PC? (dual core AMD, I think not). Second, how do you measure power ? (is this for their applications, or some other metric) If they ran the numbers they would find the cluster rather typical - unless there is more to the story.
Yes they have a a lot of processors, however, lots-o-processors != supercomputer
Cutting to the chase, did [Donald Becker] advance a "political" stand among [his] peers within the public-funded HPC community, or [was he] just trying to get some work done with the budget available at NASA?
C. None of the above. Clusters are about economics and the effect commodity hardware
has on the market. Don did what any good engineer does, "he asks what if?"
For those interested, there is a new website on clusters called ClusterMonkey. It just got started and has
plenty of good free content (and more is coming).
I got so tired of responding to these "Supercomputer
in your office" articles, I put up a web page
Cluster Urban Legends to help separate the fact from fiction.
Slashdot Mirror
Like most, I have no idea what a dangerous site looks like and I assume most people, like me, cannot tell the difference so if it is a fair test, then the majority should have had 4 right. (they were picking randomly). Disclosure of the full results should show this, otherwise it is bogas test.
What does this mean? Well, apart from the misuse of statistics for marketing purposes, I think is more of the "blaming (charging) the victim" marketing mentality. For the average users, they have been told "Here is this great thing called the Internet, go forth and enjoy the new world". Then they are told, "You are so stupid you can't tell which site will wreak your computer, you need to buy our software."
The problem needs to placed on software allows such unsanctioned downloads (i.e. the browser and OS). Of course you cannot stop the user from clicking where they should not, but with the current majority of systems running Swiss cheese software like IE users will continue to get infected.
There is no way the average user can tell if a well designed website is indeed a mal-site. Until users start demanding accountability for poorly designed software, it will continue.
- Sony's business and products
- Japanese business culture
- The English language
Yes, Microsoft might buy _______ as well. Fill in the bank with what ever you want. I thinking Victoria's Secret.Err.. I said "(for almost all applications) ". Please provide some applications that use this method. I'm not trying to troll, I have a genuine interest in heterogeneous clusters -- and how to use them efficiently for applications that currently run best on homogeneous systems.
Please list the applications you are referring to.
It is the cluster equivalent of how many people you can stuff into phone booth. If you have nothing better to do, well why not I guess.
HPC/Beowulf clusters are about building machines around problems
That is why Clusters are such a powerful paradigm. If your problem needs more processors/memory/bandwidth/data access, you can design a cluster to fit your problem and only buy what your need. In the past you had to buy a large supercomputer with lots of engineering you did not need. Designing clusters is an art, but the payoff is very good price-to-performance. I even wrote an article on Cluster Urban Legends the explains many of these issues.
The transputer was not multi-core. It could be connected to other transputers very easily, however. What it did have that made it rather interesting was process management in hardware.
If everyone has that kind of attitude, computers would be different. However, computers today are a "time sink" basically running broken software. Unless you are a specialist with adequate training, computers are a waste of most peoples time. Use a pencil and paper, calculator, and typewriter, it is more efficient. Sounds silly? How many times have you had to tell someone how do use a word processor? And how many copies did they print (waiting paper), before they got that hanging indent right (using spaces) only to find that when they change the font and the document falls apart. Need I continue...
Not that every program needs to use extra CPUs, but developers who have experienced continued speed "free lunch" improvements are going to hit a wall unless they start thinking in terms of threads, OpenMP, and MPI. You can check out Cluster Monkey for infromation on cluster computing which has been dealing with these issues for the last ten years.
I'm all for it. As long as they teach my religion as well. And by the way, although my religion is new to me, it has existed in since the beginning of time. It was created "new" only recently.
It also answers some of the more fundamental physics questions better than the bible.
This is kind of odd. Burton Smith is not really a cluster guy, although he probably knows his way around HPC (High Performance Computing). Cray is not really a cluster company (except for the system they bought from Octiga Bay deal). If you want to read a review of what Bill Gates said at the recent Supercomputing conference, check out Where is the Cluster? at Cluster Monkey.
The writer is (deliberately) confusing open source with open standards. Open standards are how progress happens. Take for example ASTM Think what would happen a a large construction company owned and wrote the ASTM building codes. This is the problem with a proprietary document format from one company. Microsot is free to implement OpenDoc, but then is will reduce their "lock in".
Stop debating creationism vs Darwinism. I want equal time (and coroprate support) for the FSM Theory (and religion). I just spent the last night working on education material for my new found religion.
The real question is whether the the PS3 will have an Linux hard disk option like the PS2. If that is the case, it may be the cheapest way to get actual development hardware.
I beleive this statement was made by a low level intern in Seoul.
Doug - a genuine Cluster Monkey
The lines have blurred due to clusters. My definition is "a collection of hardware that provides a non-trivial level of performance on a single problem" Of course, "non trivial" has various interpretations. And, working toward solving a single problem is important. Rendering is a trivial parallel application as it is really a bunch of small independent problems. Most supercomputer applications would probably run "sub-optimal" on this system (I assume it has GigE as an interconnect) because they require much more processor to processor communication. BTW, I run the ClusterMonkey site that talks about clusters and HPC if you want to learn more about clusters.
Such statements are utter nonsense. First, 15,000 PC's - what kind of PC? (dual core AMD, I think not). Second, how do you measure power ? (is this for their applications, or some other metric) If they ran the numbers they would find the cluster rather typical - unless there is more to the story.
Yes they have a a lot of processors, however, lots-o-processors != supercomputer
C. None of the above. Clusters are about economics and the effect commodity hardware has on the market. Don did what any good engineer does, "he asks what if?"
For those interested, there is a new website on clusters called ClusterMonkey. It just got started and has plenty of good free content (and more is coming).
No, I am Spartacus
Microsoft vi
Microsoft apache
Microsoft php
Microsoft emacs
I'm not trying to troll, just curious.
Take a look at The Value Cluster Project.
I got so tired of responding to these "Supercomputer in your office" articles, I put up a web page Cluster Urban Legends to help separate the fact from fiction.