A kernel level message passing service

David spoel at xray.bmc.uu.se
Fri Sep 1 04:23:35 PDT 2000


On Fri, 1 Sep 2000, Gerry Creager N5JXS wrote:

>I think, at least at first blush, that this could be a really welcome
>addition for those of us performing message passing and using scalable
>parallelism in this forum (OK, flashy words to say, "All of us.").  That
>said, I think we need to think through this and see if it would be
>something Linus could implement into a production kernel without
>forking, or would we cause a kernel fork that would cause us to lose the
>benefit of continued development, and have to reintegrate the messaging
>functionality patch into each subsequent release?

Of course this could be developed into the linux kernel as an option
during compilation. The same goes for Josep Loncaric's patches for low
latency. It need not be built into mainstream kernels, but could be a
configuration option. To be acceptable for the LK gurus it should not
introduce any security problems of course.



Groeten, David.
________________________________________________________________________
Dr. David van der Spoel		Biomedical center, Dept. of Biochemistry
s-mail:	Husargatan 3, Box 576,  75123 Uppsala, Sweden
e-mail: spoel at xray.bmc.uu.se	www: http://zorn.bmc.uu.se/~spoel
phone:	46 18 471 4205		fax: 46 18 511 755
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++






More information about the Beowulf mailing list