Clusters

A cluster is a set of one or more Compute Server nodes where optimization jobs can be performed. Each node has a job limit that indicates the maximum number of jobs that can be executed on that node simultaneously. The limit should reflect the capabilities of the machine (cores, memory, etc.) and typical job characteristics.

Compute Server nodes support a number of advanced capabilities such as job queueing and load-balancing. And if a node fails, new jobs are dispatched to other available nodes.

The Cluster Nodes Page

The Cluster Nodes page in the Cluster Manager looks like this:

nodes table structure

Job Limit

The job limit for a node determines how many jobs can be executed simultaneously on that node.
This limit can be changed using the JOB LIMIT button 5. This button opens a dialog that allows you to specify a new limit:
job limit dialog
If you submit a job to a cluster where all nodes having reached their limits, the job is added to the job queue.

Only users with SYSADMIN or ADMIN roles can change the job limit.

Stop / Restart

The Cluster Nodes page allows the user to stop or restart a node of the cluster.
Stopping the node is done through the STOP button, which is available if the node is alive. When clicked, the button opens a dialog that requests confirmation:
confirm stop node dialog
Confirm stop node dialog
After the node is stopped, the STOP button for the node is replaced by a START button, which allows the user to start the node again.

Only users with SYSADMIN or ADMIN roles can stop or restart a node.

Licenses

Access to the Licenss page is restricted to SYSADMIN and ADMIN users.

The Licenses page provides details on the licenses being used for each node of the cluster.
node licenses page
Node licenses page