Slashdot Mirror
Microsoft Competes In Supercomputer Market
HoboMaster writes "Microsoft is releasing a public beta of Windows Compute Cluster Server 2003 in their first attempt to compete in the supercomputer OS market. Gates is planned to speak at the 2005 Supercomputer Conference, which will be Microsoft's first appearance at the conference. Gates, as always, has high hopes for this new version of Windows, even claiming it to be as powerful and easier to use than Linux."
And I can bet it won't be included with their client systems.
You have two hands and one brain, so always code twice as much as you think!
Exactly. With microsoft usually charging per processor, the cost of building a super computer could go through the roof. Not to mention that they will probably build in limitations like maximum 4 processors, and to use more, you'll have to buy the enterprise edition, and spend 3 times as much per processor.
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
Does this version continue to use share nothing and thus useful be mostly for high availablity? Or can resources now be shared concurrently between different nodes of the cluster and thus provide better performance?
Obviously the comment about "easier to use" is inane when talking about supercomputers, but that quote was invented by the submitter. What the director of the HPC unit (not Gates) actually said was "...easier to integrate into what they are already doing".
What I'm listening to now on Pandora...
Yes but the issue is all the performance doesn't matter if your researchers aren't using it to solve problems. See http://www.darpa.mil/ipto/programs/hpcs/ for more info on this. This is the big HPC push that IBM, Cray, and SUN are participating in. Also a company that I think is kinda cool http://www.orionmulti.com/ is working on a very common use of HPC tools by non-computer people. They are very focused on providing easy to use ultra low maintainance computational tools primarially for the bio-informatics community. One of the founders of the company worked an LANL on green-desitiny (or something like that) which was designed to be a low power low maintainance super computing resource at LANL. After all that, the short answer is yes performance is important but there is a lot of work and interest in making sure that this performance can actually be used by the people that are actually solving problems. Mark
Ohh, I like a good GUI. Note the fact that I said "good." The very thing I love about UNIX is that you can, and probably should, do everything via a text/command interface. It's a lot faster and easier to admin large numbers of UNIX boxes then Windows, in my experience. Not to mention UNIX is quite a bit more flexible when it comes to networking and data sharing.
However, there's some things that gain great benefit from a GUI. Any sort of large scale user management, especially with a directory services type system, tends to be easier on the admin when you can visualize the directory and user account placement within a GUI.
I also like GUI based reporting - you get a much better sense of what's happening when you can see it in a chart or graph in front of you. Data visualization is good stuff. Being trapped in a GUI for administrative tasks isn't.
Windows 2000 and 2003 server have both made a lot of progress when it comes to doing things on the command line but they still fall short of UNIX because of the fact that it's non-trivial to tie the various (and completely different syntaxed) tools together. UNIX shell scripts are easy- learning windows scripting is a lot less easy and more difficult to impliment on many machines.
I'd rather start with a command-line based system and build a GUI on top of it, instead of the other way around. That way, you can always admin things 100% with ssh, customized to your organization.
- It's not the Macs I hate. It's Digg users. -
Tell that security part to the NSF, Stanford or a bunch of other
HPC cluster sites that were "owned" last year thru a security issue.
Scientific computing isn't hooked into the Internet ?? get real, it
most certainly is. Unclassified sites often support users across
the globe. For those HPC sites hooked into the Internet that want
to play fast and loose with security, be my guest. Just remember
when your funding dries up because you were "hacked" or lost
some researchers data that "security isn't an issue".
Just watch the making of documentary for epIII of star wars, and look at all the shiny G5's hooked up to xserve's with awsome apple cinema display's.
Better yet, watch the end credits - look for the huge AMD logo. Episode III was rendered on Opterons, not XServes.
You do get an reaction - the icon is selected.
You don't always want to open something when you click on it. Sometimes (mostly, I'd wager) you just want to select it for further manipulation.
Why settle for a poor interface just because it's the Windows way? (No doubt there's a way to change KDE's default behavior, and perhaps someone will explain it.)
It wasn't changed because "it's the Windows way", it was changed because it makes more sense. Indeed, Windows tried the "single click to launch" way back in 1996-1997 or so, with "Memphis" (later became IE4) and it was soundly canned by pretty much everyone who used it. KDE copying this idea was not one of its smarter moments.
I'd be interested to see if you can cite any sources for "single-click" to be "better UI" (or even just a well-reasoned argument). Microsoft would have done the reasearch back when they tried it (and Apple probably did some on the same thing way back in the early 80s), but the results are probably not easy to find.