> On Wed, 3 Mar 1999, Troy Benjegerdes wrote:
>
> >> As a side note. How difficult would it really be for someone to
> >> SQUASH MOSIX by releasing a FULLY GPL'd product that is as
> >> functional or more?
> >
> >MOSIX has existed as various extensions to various OS kernels at least as
> >early as the PDP-11 days. Their research group has been doing this for at
> >least 20 years. I doubt you could duplicate that work anytime soon.
>
> For what it is worth, UNIX itself has been around for 30 years,
> and now we have Linux. I'd bet that should someone here decide
> to reimplement MOSIX or clone it, that they could do it, and it
> wouldn't take 20 years. I'd bet that if someone were as
> committed as Linus was creating Linux, that they could start such
> a project, and get enough followers, coders, and hackers to whip
> out a full clone/replacement of MOSIX before a year or 2 expired.
I would expect this to take more like 5 years. Sure, someone could whip
out a clone that can migrate processes and look like it's doing load
balancing, but the performance issues in parrallel computing are so much
harder to debug, and even quantify. Writing a MOSIX clone is going to take
a great deal of knowledge, not to mention a cluster to test on.
I propose that those of you who want a MOSIX clone, go get the GPL'ed
kernel modifications they made, and us it as a base for a generic
load balancing and process migration API for the linux kernel.
I would hope that this idea could then resolve this license fiasco, and
allow those of us that are interested in using/evalutating MOSIX to do so,
and provide a framework for future GPL'ed work.
--------------------------------------------------------------------------
| Troy Benjegerdes | troy@microux.com | hozer@drgw.net |
| Unix is user friendly... You just have to be friendly to it first. |
| This message composed with 100% free software. http://www.gnu.org |
--------------------------------------------------------------------------
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/