Messages in this thread | | | Subject | Re: Help in DSM design | Date | Sun, 05 Mar 2000 14:24:17 -0800 | From | Larry McVoy <> |
| |
: Obviously DSM works for some situations and horribly breaks : in others. What about sticking non-intrusive DSM code in the : kernel and attach a BIG FAT WARNING to it? : : People don't have to use it; as long as the DSM code doesn't : impact a performance of a non-DSM kernel, I really don't see : a problem.
This is the same argument that managers at Sun, HP, SGI, IBM, and the company formly known as DEC used. There are a million and one things you could ask for in the kernel and they are just like cache line misses. Nobody notices them when you look at them one at a time but 5 years later you wake up and your kernel is full of misguided, half finished, poorly thought out ideas that make it impossible to do anything worthwhile.
It's a lot like your diet. One beer isn't going to hurt you, but to use the "one beer isn't going to hurt you" argument to justify drinking a zillion different beers is not smart.
So why pick on DSM? Because of all the misguided, half finished, poorly thought out ideas out there, this one is one of the most studied and proven mistakes in the history of computing. If we can't keep this sort of thing out of the kernel, we might as well pack it up and go run Solaris.
- 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/
| |