Product specifications

Each time that Connection Server requires information, the data access driver opens a connection
to the database, retrieves the data, then closes the connection.
Connection Server keep available connections open and maintain their details in a connection pool.
Each time that Connection Server requires information from the data source, the data access driver
checks the connection pool to see if it contains an unused, suitable connection. If an existing
connection is available, the connection is used. If all connections are being used, Connection Server
creates a new connection and adds it to the pool. This method uses system resources more efficiently.
4.4.1 About connections of the pool
Caution:
This section only relates to JDBC connections.
Connections that are available in the connection pool can be exclusive or shareable.
Exclusive connections can be allocated to one user only at a time. When an exclusive connection
is allocated, it is no longer available in the pool. Then it cannot be allocated to any other requesters.
When the connection is no longer required, the custom driver releases the connection so that it can
be reallocated.
Shareable connections can be allocated to multiple users at a time. When a connection is allocated,
the connection remains in the pool so that it is available for other requesters.
Related Topics
Connection Shareable
Shared Connection
Max Pool Time
4.5 Load balancing
In server mode, you can run several Connection Server instances on the BI platform. Connection Server
implements a load-balancing mechanism in this deployment mode. It chooses the best server in term
of resources that can support new client requests.
Load balancing helps to address scalability issues by increasing the number of server instances on the
BI platform.
Caution:
This load-balancing mechanism is different from the load balancing implemented by the Central
Management Server.
Load balancing is performed at the following levels:
2012-03-1532
Learning data access specifics