Load Balancing in Cloud Computing Environment Using Improved Weighted Round Robin Algorithm for Nonpreemptive Dependent Tasks

Table 1

Cloud setup configuration details.

Sl. number	Entity	Quantity	Purpose
1	Data center	1	Data center having the physical hosts in the test environment
2	Number of hosts in DC	500 hosts (200 Nos-4-CoreHyperThreadedOpteron270, 200 Nos-4-Core HyperThreadedOpteron2218 and 100 Nos 8-Core HyperThreaded XeonE5430)	Number of physical hosts used in the experiment
3	Number of process elements	8/8/16	Number of executing elements in each of the hosts. The host has 8 or 16 processing elements
4	PE processing capacity	174/247/355 MIPS	Each host has any one of the processing capacities
5	Host ram capacity	16/32 GB	Each host has any of these RAM memories
6	Number of VM	10 to 100 with an increment of 10	Number of virtual machines used in the experiment
7	Number of PE to VM	1	Processing element allotted in each VM
8	VM’s PE processing capacity	150/300/90/120/93/112/105/225	Virtual machine’s processing capacity
9	VM RAM capacity	1920 MB	The RAM’s memory capacity of the VM
10	VM manager	Xen	The operating system runs on the physical machine to manage the VMs. It provides the virtualization
11	Number of PE in Tasks	1	The job/task’s maximum usable processing element
12	Task length/instructions	500000 to 200000000	Tasks length in million instructions. The heterogeneous job length test having the variations from the mentioned minimum to maximum