There are two types of Jobs that can be submitted to a job queue server: A single GeoDict command or a GeoPy Macro.

Submitting a Single GeoDict Command

Most GeoDict commands can be run in the job queue. To do so, select the desired module (e.g., FiberGeo). In the module section, select the desired functionality and edit the settings as you normally would do when running simulations locally. Instead of clicking the Run/Generate button to run the command locally, click the Job Queue button.

Run Command in GeoDict Job-Queue

The Run Command in GeoDict Job-Queue dialog opens, where you can define multiple settings:

• Project Name: This name allows you to group multiple simulations together by giving them a shared project name. The default name comes from the name of the project folder that is currently selected, e.g., MyFirstGeoDictProject.
• Job Name Prefix: This name serves to identify the specific simulation that is run. The default name depends on the chosen command, e.g., FiberGeoCreate for creating fibers using FiberGeo. If there are multiple jobs with the same job name prefix within one project, the name is extended by a running number (*-001, *-002, ...).

• Upload Current Structure: When this option is activated, the currently loaded structure in GeoDict is uploaded to the job queue server along with the job. This is necessary for modules that analyze structures or predict properties. This option is active by default when a structure is loaded in GeoDict.
• Upload Current Project folder: When this option is activated, the entire current project folder, along with the job, is uploaded to the job queue server. This is necessary if additional files in the project folder are required to run the computation. For example, this would be necessary to continue a FlowDict simulation from a previous result.
• Selected machine: Here, you can select the job queue server you want to use for the simulation. The last server used is selected by default. If only one server is available, it will be selected automatically. Click Select to open the Select server for job dialog. Here you can see all currently reachable job queue servers. For each server find the following information:
  
  • The Hostname of the machine.
  • Your Username on this machine.
  • The number of Max. parallel jobs that is allowed to run simultaneously on this server. The server admin can change this value via the Job Queue Settings.
  • The number of currently Running jobs.
  • The number of Jobs in queue waiting to be started.
  • The available Number of logical cores and Total RAM together with their current usage CPU utilization and RAM in use.

• Timeout: This option allows you to set a timeout for the job. After the defined number of hours, the job will be terminated automatically. If you do not want to set a timeout, set this value to 0, and the simulation will run until it finishes, is terminated, or crashes. Setting a timeout helps prevent the job queue server from becoming blocked by jobs that encounter convergence problems, for example.

After defining all settings, click on Run to send the job to the server. A dialog showing the progress of Packing, uploading and starting the job is displayed. For each job a python macro, based on the selected command and its settings is created and then executed on the job queue server. After the upload of the job is completed the Job Queue Projects dialog opens automatically. There, you can see the current status of the job and download the results of completed jobs.

Submitting a GeoPy Macro

You can also run GeoPython macros and scripts on a job queue server. To do so, open the Macro Execution Control dialog, select your macro, and click Run in Job Queue. This is especially useful for macros where parameters can be varied by selecting Vary. Define multiple values for the corresponding parameters. Then, for each parameter combination of the variation a single job is created and queued on the server.

After clicking on Run in Job Queue the Run GeoPython Macro in GeoDict Job-Queue dialog opens, which is very similar to the dialog for running a single command:

• Project Name: This name allows you to group multiple simulations together by giving them a shared project name. The default name comes from the name of the currently selected project folder, e.g., MyFirstGeoDictProject.
• Job Name Prefix: This name serves to identify the specific computation that is run. The default name consists of Macro execution and the file name of the macro. This file name is extended by the parameter values, e.g., Macro execution Tutorial-D-GeoPy-Beginners-Workshop gd_SVP=5 gd_FD=5.

• Upload Current Structure: When this option is activated, the currently loaded structure in GeoDict is uploaded to the job queue server along with the job. This is necessary for modules that analyze structures or predict properties. This option is active by default when a structure is loaded in GeoDict.
• Upload Current Project folder: When this option is activated, the entire current project folder, along with the job, is uploaded to the job queue server. This is necessary if additional files in the project folder are required to run the computation. For example, this would be necessary to continue a FlowDict simulation from a previous result.
• Selected machine: Here, you can select the job queue server you want to use for the simulation. The last server used is selected by default. If only one server is available, it will be selected automatically. Click Select to open the Select server for job dialog. Here you can see all currently reachable job queue servers. For each server find the following information:
  
  • The Hostname of the machine.
  • Your Username on this machine.
  • The number of Max. parallel jobs that is allowed to run simultaneously on this server. The server admin can change this value via the Job Queue Settings.
  • The number of currently Running jobs.
  • The number of Jobs in queue waiting to be started.
  • The available Number of logical cores and Total RAM together with their current usage CPU utilization and RAM in use.

• Timeout: This option allows you to set a timeout for the job. After the defined number of hours, the job will be terminated automatically. If you do not want to set a timeout, set this value to 0, and the simulation will run until it finishes, is terminated, or crashes. Setting a timeout helps prevent the job queue server from becoming blocked by jobs that encounter convergence problems, for example.

• Variable Variation: When a macro that uses Vary for the parameters is selected, a table containing the varied parameters and all of their combinations is displayed at the bottom of the dialog box. Each combination is submitted as an individual job; thus, nine jobs are submitted in the example shown.