Process pooling is available in the following three modes:

• Built-in pooling
• AppServer pooling
• Disabled pooling

Using built-in pooling, OpenEdge Management handles a set of OpenEdge client processes. When the ABL APIs are called, these processes improve performance by eliminating the overhead of starting and stopping a process each time. You can control the maximum number of _progress processes that are kept alive at any time. The processes do not maintain a persistent connection to the database. Instead, each request to OpenEdge Management from a web browser causes one of the AVMs (_progress.exe) to connect to a database and then run the proper data administration API. Built-in pooling is the default process pooling mode.

Using AppServer pooling, it is the AppServer, instead of the OpenEdge client process, that runs the ABL APIs. Each request from OpenEdge Management is sent through the specified AppServer using a specified AppServer URL to call the ABL API. The Java Open Client is used to connect to the specified AppServer.

When you disable pooling, you use the least amount of memory. However, you also experience the least effective performance. A new AVM is started on each request from the Database Administration Console. A new database connection is made on the startup of each OpenEdge client process (_progress), and the ABL API is run using the -p startup parameter.

You can disable pooling in the following cases: