Messages in this thread | | | From | (Larry McVoy) | Subject | Re: Lets get this right (WAS RE:MOSIX and kernel mods) | Date | Sat, 06 Mar 1999 10:28:37 -0800 |
| |
From "Albert D. Cahalan" <acahalan@cs.uml.edu> : > We need a solution that we can live with for the next ten years, not : > one that works now, and then we have to ditch at the next major : > release with much gnashing of teeth. : : FUD, FUD, FUD. (Why?) Distributed memory won't need to be ditched.
I've got a question for you, since you seem to be implying that the people saying "no thanks" are spreading FUD, and that you actually know what you are talking about. Please answer these questions or just shut up and go away:
. Please list the research and practice papers on DSM that you have read, and summarize their conclusions . Please describe your experience with systems similar to that which you are suggesting be included with Linux. List the number and types of applications which worked on these systems, as well as the apps which did _not_ work on these systems. . Please tell us the locking rate in your applications. Describe the throughput rate of your apps as a f(# of nodes, lock latency). Show the curve.
: Let's say you have 3 choices: : : 1. 95% use of a single machine : 2. 80% use of a 16-way cluster, but 2 weeks development time : 3. 60% use of a 16-way cluster, but 2 hours development time
Show me the apps which match this model. Where are they? Remember, you have to come up with an app that you can get the 60% out of the cluster using DSM and you have 2 hours to hack it to work. Demo programs don't count, a real app, please. And if I can hack it to usin PVM or MPI in the same or less time and get the same or better performance, then you lose.
: It is simply cruel to force your idea of perfection on people who : have completely different needs than you do.
Demonstrate those needs and show the math which say your solution works. otherwise it is you who is spreading FUD, not the other guy.
- 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/
| |