Play tic-tac-toe? "Stojanovic has lost to MAYA more than a 100 times." With semi-intelligent players I thought this game was pretty much guaranteed to generate a draw?
There wasn't time to patch all of our servers during the outbreak, so one of the guys here implemented a group policy that prevents execution of msblast.exe and teekids.exe on any machine on our network.
There was time to patch before the outbreak and there have been advisories for weeks that the worm was coming. This guy would have been smarter to apply the patches in the first place.
500 cameras, say at least 10KB/sec per camera, that's 5 MB/sec, 18 GB/hour, at least 8 hours a day, so about 150GB a day. About 200 days in a school year, 30 tera bytes/year.
I can't say where the numbers came from, but it wasn't from marketing. What percent of peak you get depends heavily on what problem you're trying to solve. I don't happen to know the problem(s) that were getting 10% of peak and calling it good, but I suspect they were high bandwith problems.
To simply reassert your "10% of peak" without support adds nothing to the discussion. Lots of Beowulf clusters get 30 to 70+ percent of peak on lots of real world problems, you can find some via Google quickly.
I've heard (no, I don't have actual links to articles) that 10% of peak performance on a cluster is considered really good.
Sounds like Cray marketing articles. For example, Daniel Katz at JPL wrote in 1997:
it is possible to construct a 16-node machine with a theoretical peak performance of 3.2 GFlop/s and a typical sustained performance of 1.2 GFlop/s
which is > 35% of peak. Or consider this from the Universiry of Liverpool:
The current Beowulf cluster can deliver a theoretical peak performance of about 100 Gigaflops (billions of floating point operations per second) and has been observed to deliver about 60 Gigaflops.
The observed performance was based on LU decomposition.
For sustained/peak of about 60%.
I have no doubt that one could find problems where a Beowulf cluster has 10% efficiency, but there are real many problems that are good to go on a cluster. And even if you only got 10% it would be worth it if the cluster cost 5% of what a vector computer costs. Not to mention that performance/$ on commodity hardware increases by a factor of 2 every 12-24 months. It takes years to develop a supercomputer, and they are stuck at their level of technology for several years since they are so expensive to redesign.
Now, that being said, I would argue that Perl 5 already presents 99% of Lisp's flexibility.
I disagree. Much of Lisp's flexibility comes from having programs represented in the same form as data (S-expressions) which lets you build a data structure and then pass it to one of Lisp's EVAL family. Perl's eval(), while useful, is a poor cousin of Lisp's facilities.
The tags will only cost about $25 and I am sure that price will go down w/ time.
From the article:
The tags -- expected to cost less than $25 each to produce
$25 is the approximate cost to produce - they will no doubt cost more to buy.
It used to bother me that/. posters didn't RTFA but I'm starting to realize that even when they do read the articles many don't understand what they've read before they post.
But looking again, you can get a PIII 1G and the motherboard as a combo for $65.
You can get a case for about $30. You can get the video card for about $80.
So a better system for cheaper
You left out the CD/DVD reader - which the XBOX has - that pushes your homebrew box over the $200 mark.
That way, they could switch themselves over to the best offerings from Intel and/or AMD at a moment's notice rather than get caught with sub 2 GHz G4's for the next 2-3 years until a rushed port is available. Or worse yet, Motorola decides to sell off or get out of the CPU business.
not to mention that MS basically partered with IBM on OS/2, then back-stabbed them while secretly working on a competeting OS (windows).
MS did not work on Windows in secret. Although most people think that the history of Windows starts with 3, there were earlier versions. They pretty much sucked, and not many apps were written for them. However at one point Excel was supported on OS/2, Windows version 2 (yes, there was a Windows 2), and Mac.
Windows 3 was considered by many to be a stopgap until a new OS/2 was ready but turned out to be popular, and the strategy shifted from moving from Windows toward OS/2 to a strategy of expanding on Windows.
I attended a talk given by Jim Gray on the subject of "Databases Meet Astronomy" about a year ago. He gives a lot of talks on databases and science. He talked about sky surveys generating petabytes of data. The VLBA radio observatory generates 1 gigabyte per second. Much of the data mining could be accomplished with a google type model of lots of machines working on pieces of the problem.
He also talked about CERN generating 10 PetaBytes a year when their new collider comes on line
Supercomputers are sexy, but are losing the technology war. If you start designing a new one today it will be years before it is ready. During those years Intel and AMD will crank up their clock speeds and negate much if not all of the CPU speed advantage you get from your fancy design. Why not go for parallelism from cheap machines?
Slashdot Mirror
I imagine he lost in order to test the system.
Ah, the Smartphone that Sendo is suing Microsoft over.
How many companies or non-profit organizations have their own Space Shuttles?
There was time to patch before the outbreak and there have been advisories for weeks that the worm was coming. This guy would have been smarter to apply the patches in the first place.
500 cameras, say at least 10KB/sec per camera, that's 5 MB/sec, 18 GB/hour, at least 8 hours a day, so about 150GB a day. About 200 days in a school year, 30 tera bytes/year.
Sounds like Cray marketing articles. For example, Daniel Katz at JPL wrote in 1997:
which is > 35% of peak. Or consider this from the Universiry of Liverpool:For sustained/peak of about 60%.
I have no doubt that one could find problems where a Beowulf cluster has 10% efficiency, but there are real many problems that are good to go on a cluster. And even if you only got 10% it would be worth it if the cluster cost 5% of what a vector computer costs. Not to mention that performance/$ on commodity hardware increases by a factor of 2 every 12-24 months. It takes years to develop a supercomputer, and they are stuck at their level of technology for several years since they are so expensive to redesign.
I disagree. Much of Lisp's flexibility comes from having programs represented in the same form as data (S-expressions) which lets you build a data structure and then pass it to one of Lisp's EVAL family. Perl's eval(), while useful, is a poor cousin of Lisp's facilities.
Execute data, not code.
Has Linux been ported to the PSP yet?
Ranges of a few miles can be had with Pringles(tm) can antennas.
It only uses power when it is on. You could turn 802.11b on only when you want to upload pictures.
They're having fun. They'll laugh a lot the rest of the day. They'll tell a whole lot of people about it. That's not doing nothing. Works for me.
Will it be released at the same time as HURD?
When I got to the page with the article there was a big ad for Windows Server 2003.
LARGER!!!
add THREE
INCHES to
your styl
us!!
The tags will only cost about $25 and I am sure that price will go down w/ time.
From the article:
The tags -- expected to cost less than $25 each to produce
$25 is the approximate cost to produce - they will no doubt cost more to buy.
It used to bother me that /. posters didn't RTFA but I'm starting to realize that even when they do read the articles many don't understand what they've read before they post.
You left out the CD/DVD reader - which the XBOX has - that pushes your homebrew box over the $200 mark.
So you didn't notice that Apple announced that they'll be using an IBM processor in their next PowerMac desktop line?
MS did not work on Windows in secret. Although most people think that the history of Windows starts with 3, there were earlier versions. They pretty much sucked, and not many apps were written for them. However at one point Excel was supported on OS/2, Windows version 2 (yes, there was a Windows 2), and Mac.
Windows 3 was considered by many to be a stopgap until a new OS/2 was ready but turned out to be popular, and the strategy shifted from moving from Windows toward OS/2 to a strategy of expanding on Windows.
I thought we were supposed to refer to Linux as GNU/Linux. If so, then when Linux is attacked, so is GNU!
Sorry, wrong. QNX USB support.
Space isn't two dimensional (square mile) - the measure would have to be a certain number of molecules per cubic mile.
Intel has previously shipped over a billion of their 8051 8 bit microcontroller.
He also talked about CERN generating 10 PetaBytes a year when their new collider comes on line
Supercomputers are sexy, but are losing the technology war. If you start designing a new one today it will be years before it is ready. During those years Intel and AMD will crank up their clock speeds and negate much if not all of the CPU speed advantage you get from your fancy design. Why not go for parallelism from cheap machines?