Our objective is to develop an efficient load balancing algorithm like throughput, la-tency, Fault tolerance for example for the clouds of different sizes like virtual topology counting on the appliance requirement. Cloud computing involves virtualization, distributed system and some web services. A cloud consists of several elements like clients, data centre and distributed servers. Central to those issues lies the establishment of an efficient load balancing algorithm. The load is often CPU load, memory capacity, delay or network load. Load balancing is the process of distributing the load among the various servers of a system to enhance both resource utilization and job reaction time while avoiding a situation where many of the servers are heavily loaded while other servers are idle or doing little or no work. Load balancing gives each one within the system or every server within the network does approximately the equal amount of labour at any instant of the given time. These techniques are often sender initiatives, receiver initiatives or symmetric type like combination of sender initiated and receiver initiated types.