|
Hello,
most of you said, DON'T do it this way. o.K. :-(
I only want to mention a last thing.
Our database is not a relational Database.
Our problem is much more like a Meyers Lexica.
The first book goes from "A" to "Be", the second
from "Bf" to "C", etc.
If you want to lookup for Example "Aircraft", you know
its in the first book and don't have to look in any other.
Thats why we DON'T need shared memory, and why
MOSIX or Beowulf or any other "real" cluster system is
not the right thing for us.
We want the scheduler to analyse the first two byte of the
request, and then send it to the corresponding real server.
Sounds that better ??
Thank you
mbernhardt.vcf
Description: Card for Markus Bernhardt
----------------------------------------------------------------------
LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
To unsubscribe, e-mail: lvs-users-unsubscribe@xxxxxxxxxxxxxxxxxxxxxx
For additional commands, e-mail: lvs-users-help@xxxxxxxxxxxxxxxxxxxxxx
|
