Hi all!
I've been a lurker for quite a long time in this list :)
Currently I am planning on doing my final project for my bachelor degree.
I am interested in encanching load balancing policy on LVS, after googling
and search thru this list archives i've found that a neural network based
load balancing is still an interesting topic for load balancing.
IMHO a neural network based load balancer is an intelligent load balancer
which can predict the performance of each cluster and then select the
appropiate server to handle an upcoming request.
For example take the respose time of each server (or other load metrics
like cpu, memory, disk and network load) as input for neural network and
then load balancer learn from previous response time of each server and
predict which server will be suitable for next request.
Any sugestion or idea? Has somebody already working on it?
Please give me a hint ...
Sorry for my English
TIA
Asep Noor MS
Student of Department of Computer Science
Bogor Agriculture University, Indonesia
|