A load balancing model based on cloud partitioning for the public cloud

Question

i am trying to do a project using the ieee paper "A load balancing model based on cloud partitioning for the public cloud" i am using java for coding but i don't know where to start i don't know anything about cloud computing. the algorithm described in the paper is like there is a controller and when ever a new process is requested the controller will search for nodes which has less load and assign that process to that node.

is it possible to measure the load of each system in a network and how can i assign each process to these nodes

Answer 1

Do you have a setup of hypervisors / cluster ? If you have the physical setup ready then start setting up Open Stack . Java alone can't help you much, you need to know Python. Learn how the provisioning is happening in Open Stack . Now Open Stack has different modules and I guess Load Balancing As a Service is implemented for the network component neutron. You can probably take the help from here and can think how you can write a load balancer for the nodes while provisioning.

I donno but this may help you !

Answer 2

The nodes load is calculated by cpu load, ram usage, network usage etc. All these parameters are obtained from the system just like the task manager which shows the details of the cpu usage and others. Search for the code of cpu usage, ram usage etc on net and calculate load degree.