>A while back someone brought up the idea of modifying the `make -j N'
 >usage to incorporate support for distributed makes across multiple
 >machines.  So maybe the syntax would be something like:
 >
 >  make -J hostname1[,hostname2,...]
 >
 >You'd have to have the hosts configured properly to handle remote logins,
 >but I think this should be a workable (and extremely cool) modification.
 >
 >This would eliminate the need to manually subdivide which bits of the file
 >tree are arbitrated to compilation on certain hosts.


>There is actually a perl script that will do this for you. It may be >worth
>finding it instead of mucking with the kernel. I beleive it uses rsh >and
>similar tools to copy source to other machines and compile them
>automagically. I can't quite seem to find it, though. A little help?
>
>--Doug

I have thought of this too, I rather have a native make program that does 
that, doing this doesn't require any mucking with the kernel, instead of 
command line options, how about giving make a file which contains the 
hostname of the systems it will use, for a very big project ie (windows NT 
src code :P ), you will prolly want to use 10,000 systems and that will be a 
lot for commandline,   What I have been thinking about is what if there are 
depends, what if we need to build a library midway before the rest of 
compilations can continue, then spliting the file will not work.    The 
other systems will have to wait for the library to be built before 
continuing, how will the know that the library that the are waiting for is 
built?  the must have to communicate somehow, thus this is not a simple 
problem :|  A good library for distributed computing is the MPI library or 
PVM, but I am lazy, clueless and such, so I dobut I will even try to 
implement such stuff.

Segmond




_______________________________________________________________
Get Free Email and Do More On The Web. Visit http://www.msn.com