Manage the Creation and Modification of Orchestration Jobs

The creation and modification of orchestration jobs is managed by the SAS Visual Process Orchestration Design Server. The Design Server is implemented as a load-balanced server-side pool. This architecture is described in Load Balancing and Pooling chapter of the SAS Intelligence Platform: Application Server Administration Guide.

The Design Server consists of a pool of server instances that is managed by an object spawner. The object spawner listens for requests at the Design Server port. When a user opens the SAS Visual Process Orchestration Web Client, the client sends a connection request to the Design Server port. The object spawner responds by receiving authentication and connecting the web client to an available server instance.

When the web client breaks the connection, the server instance is returned to the pool for assignment to another client. If the server instance remains inactive for a specified time period, it is stopped by the object spawner.

To configure the Design Server, set the following options in SAS Management Console. Open the Server Manager and select Properties from the Design Server’s context menu.

Option	Default Value	Description
Logical Design Server Properties		In SAS Management Console, open the Server Manager and select Properties at the top level of the Design Server. Then select the Load Balancing tab.
Load-balancing algorithm	Most Recently Used	Minimizes the number of active server instances.
Availability timeout	60 seconds	Server instance request fails if not assigned within the specified time period.
Design Server Properties		In SAS Management Console, open the Server Manager and select Properties at the lower level of the Design Server. Then select Options Advanced Options Load Balancing.
Launch timeout	60 seconds	See Availability timeout.
Recycle activation limit	0 connections	Specifies a number of client connections that are allowed of a given server instance before that instance is recycled (disconnected.)
Service process maximum	50 server instances	New client connection requests are refused when the Design Server is running the specified maximum number of server instances. Requests are accepted again when the number of active server instances drops below the maximum.
Server process minimum	0 server instances	The object spawner continues to start new server instances in response to new connection requests from clients, until the minimum value is reached. Above the minimum value, the object spawner seeks to reuse existing server instances in the pool.
Inactive process shutdown timeout	600 seconds	Specifies the length of time that a server instance can remain inactive before it is terminated by the object spawner.