Vote count:
0
Describe how in the general case of n servers, each capable of serving s requests per second, you would allocate x requests from the queue to achieve load balancing. State any assumptions you make.
I am trying to give a reasonable answer to this problem. I was thinking that the simplest way to do it would be to allocate requests proportionally given the number of requests each server can process per second. If we have A (18/s), B(26/s) and C(56/s), my answer would be to pass 18 to A, 26 to B and 56 to C. In this way we would be able to process 100 requests per second. I was wondering if my reasoning was making any sense. What do you think?
asked 35 secs ago
Aucun commentaire:
Enregistrer un commentaire