Messages in this thread | | | Date | Fri, 5 Mar 1999 14:32:35 -0600 (CST) | From | Kamran Karimi <> | Subject | Re: MOSIX and kernel mods. |
| |
On Fri, 5 Mar 1999, Richard Gooch wrote:
> But I can leave that aside, since it's your code, even though I > disagree with it's appropriateness. What I strongly object to is the > bloat that is added to the kernel to support your model. If DIPC was > limited to extending message queues to the cluster, I would be less > concerned (although I feel it should be possible from user space).
As I have said before, DIPC is a config time option. You can decide if you want it in your compiled kernel. What you say is like claiming it is not a good idea to have all those driver sources in the Linux kernel just because you won't be using most of them. Please think about this.
> An MPI programme is also MP-ready, but doesn't encourage constructs > which are heavily penalised in a cluster. Thus MPI programmes written > for SMP or DM machines (with fast interprocessor connections) should > scale well to a cluster.
Yeah, but that works backwards: Creating a message passing layer on top of a shared memory architecture. Shared memory is more fundamental. IMO an appropriate "abstraction" should go the other way.
> Again: I'm all for abstractions that make life easier. But DSM makes > people think the *wrong way* about distributed computing.
I'm repeating myself: Not everybody cares!
> And CS is littered with endless "ideas" which were dropped because > they didn't work in real life. People moved on from those > discussions. It cuts both ways.
I don't think continuing this discussion will convince any of us. Good luck with your work.
-Kamran
- 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/
| |