<html><head><meta name="qrichtext" content="1" /></head><body style="font-size:9pt;font-family:Sans Serif">
<p>So, does the openMosix 2.6 development branch function correctly? I understand it's not considered <span style="font-style:italic">stable</span> as of yet, but for my purposes will it work? I supposed I could use 2.4, I'd prefer 2.6 though. Do you have any other suggestions? Would running an Xserver still be limited to one CPU though? Like if I wanted to run icewm, it would be limited to a max of 166Mhz, but firefox inside of icewm would also have ~166Mhz? Is that correct?</p>
<p></p>
<p></p>
<p></p>
<p></p>
<p>On Monday 14 November 2005 07:20 am, you wrote:</p>
<p>> On Mon, 14 Nov 2005, Jake Thebault-Spieker wrote:</p>
<p>> > I've heard of many of these(Scyld, OpenMosix, etc.) Since this is how it</p>
<p>> > will work, will I be able to use these 6 machines as a workstation? I</p>
<p>> > understand that the CPU power will be next to none, and I have read your</p>
<p>> > book about overhead, and feasability and so forth. Having not done the</p>
<p>> > calculations, I have a pretty good feeling that my computers will not</p>
<p>> > meet the standards for them to actually be useful. I'm a high school</p>
<p>> > student, and would like a multi-CPU machine to play w/, but I don't want</p>
<p>> > to spend the money. Thanks again.</p>
<p>></p>
<p>> To use the six machines "as a workstation" will probably require</p>
<p>> OpenMosix -- that permits you to login to one machine's console and just</p>
<p>> run jobs. Those jobs will then automigrate onto nodes, with network</p>
<p>> socket connections established to pipe e.g. stdin, stdout, open file I/O</p>
<p>> back to the master node transparently. For embarrassingly parallel</p>
<p>> tasks -- independent runs of some background job -- this will probably</p>
<p>> be "reasonably" efficient. For graphics intensive tasks, for tasks with</p>
<p>> a lot of I/O, it probably will be less efficient or even INefficient --</p>
<p>> you add network overhead on top of the already nontrivial I/O overhead</p>
<p>> and a different form of virtual parallelism might be called for that</p>
<p>> permits the usage of local disk resources (for example) rather than</p>
<p>> always writing to a virtualized local disk that then is socket-forwarded</p>
<p>> back to the master host and turned in to a real disk write at that end,</p>
<p>> especially if the master is running many copies of a similar/identical</p>
<p>> task so that its disk AND network subsystems are constantly colliding</p>
<p>> and nodes are having to wait in line (blocking the calling application</p>
<p>> in the meantime).</p>
<p>></p>
<p>> So sure, give OM a try. If nothing else, it will be a learning</p>
<p>> experience for you -- don't worry too much about whether or not the</p>
<p>> computers will be "useful" per se or whether you could do better to just</p>
<p>> buy a new AMD-64 motherboard unless or until your task is worth</p>
<p>> real money. Otherwise "useful" is defined mostly by whether or not you</p>
<p>> have fun with it, how much you learn from doing it.</p>
<p>></p>
<p>> rgb</p>
<p></p>
<p>-- </p>
<p>Numbers rule the Universe.
</p>
<p> --The Pythagoreans
</p>
<p>
</p>
<p>Carpe Aptenodytes(Seize the Penguins),
</p>
<p>Jake Thebault-Spieker
</p>
<p></p>
</body></html>