The process of routing user requests to application instances residing in multiple clouds can be termed as Cloud Balancing. This is also considered to be a logical extension of Global Server Load Balancing (GSLB) which ensures that all servers are loaded equally and overloading of particular server due to specific hosted application can be avoided. The ultimate aim is to ensure high availability and maximum performance of hosted applications.
As stated earlier, cloud balancing routes user requests to different clouds. The choice is made in real-time based on a number of variables. These include any or a combination of these variables - service-level agreements, business metrics, response time, capacity of a particular server, power consumption and variations, user profile and the location from which the user accesses the applications.
Cloud balancing helps availability of a particular application to varying traffic and enables uninterrupted and efficient data access. The client can do cloud balancing by hosting the application with different service providers and installing appropriate routers; it can also be done by the service provider by hosting the application across independent locations based on the requirement.
It is not as simple as it has been explained, but requires an intelligent, policy-driven and interpretive method of routing application requests. This definitely improves the performance of the hosted application, which adds significant value to the organization.