If you provide multiple AppServer instances to service client requests for an application service, the NameServer required to manage these requests maintains a list of the available AppServer instances on the network that support the given application service. When a client session binds to an application service, OpenEdge uses the list of AppServer instances provided by the NameServer to establish a connection pool on the client’s behalf.

A connection pool is created in a client session and includes a list of physical connections to AppServer instances that support the same application service. So, when a client binds to the application service, OpenEdge requests connections to AppServer instances that support that application service and uses these connections to service the client requests. The more AppServer instances and agents that are available to support the connection, the more the application service can service client requests in a truly parallel fashion. OpenEdge maintains this connection pool throughout the life of the logical connection, adding and trimming connections as required to satisfy specified client requirements given the AppServer resources that are actually available. With the load balancing option provided by the NameServer, you can also have more client traffic directed to AppServer instances running with more agents and more powerful computing resources.

For more information on NameServers and how they support logical connections to application services, see the information on AppServer run-time components and operation in OpenEdge Application Server: Administration and the information on session models and application services in OpenEdge Application Server: Developing AppServer Applications. For more information on how OpenEdge maintains logical connections for client applications, see Connection process.