Research Article
Dynamic Q-Learning-Based Optimized Load Balancing Technique in Cloud
Algorithm 2
Reward apply for VM minimum configuration.
| | If i = = master, then | | | Else If i = = slave or older, then | | | Check the total value = max value | | | Else If i = = member, then | | | End if | | | For Progress the swarm to acquire new solutions | | | If Nkill = 0 and Ns < Nsmax then | | | Make fresh VM per minimum config | | | Until X < Xmax | | | Else | | | Update SC = SC + 1 | | | If SC = SCcmax then | | | Reset the SC | | | End if | | | End if | | | Else if Nkill ≠ 0 and X < Xmin | | | Delete the final VM | | | Update the solution set | | | End if | | | While recurrence the evolving process until no new RIN learning theory | | | End for | | | IF VM load > Host_Intial | | | for every HVm < -Fact increased | | | HVm < -Newer.HostId | | | end IF | | | end For | | | Start the VM on this host | | | Start VM VM_id, LHost | | | end loop |
|