The more successful a distributed application, the higher the load that the growing number of users places on all components of the application. Often, even the computing power of the fastest hardware is not enough to keep up with the user demand.
An inevitable option at this point is the distribution of the load among multiple server machines. Section 0, "Scalability," mentions briefly how DCOM facilitates different techniques of load balancing: parallel deployment, isolating critical components, and pipelining of sequential processes.
"Load balancing" is a widely used term that describes a whole set of related techniques. DCOM does not transparently provide load balancing in all its different meanings, but it does make it easy to implement different types of load balancing.