Batches

A batch is a specification for an optimization task. It contains all inputs necessary to perform the task (model data and optimization parameters), as well as a description of the results that are needed upon completion. A batch specification is built off-line and submitted to the Cluster Manager. The Cluster Manager then submits the batch job for execution, at which point the optimization model will be solved and results will be generated. If a batch job fails for reasons outside of its control (e.g., the node it was running on went down), a new job will be created to retry the batch. Once the batch has completed, the results are stored on the Cluster Manager and a client can retrieve them.

The Batches Page

The Batches page in the Cluster Manager looks like this:

jobs table structure

Filtering and Searching

Batches displayed in the table can be filtered using the Filter box and the Search box.

Filtering

The filter box provides the user with a set of predefined filters as shown below:
Batch filter dialog

By default, users having one of the roles ADMIN, SYSADMIN, or READONLY will see all batches. Other users will only see batches they have launched. This default filter can be changed from the Filter box.

When the table has been filtered, two numbers are displayed to the right of the page title showing respectively the number of batches matching the current filtering and the total number of batches. If no batches have been filtered, only the total number of batches wil be displayed.

Searching

The Search box enables users to further refine the results by specifying an arbitrary search string. In order for a batch to be displayed, the search string must be present in at least one of the columns in table for that batch, or in the job batch ID.

It is important to note that filtering with the Filter box is usually applied server side, while searching with the Search box is always performed client side. That means that the searching is done on results that were already filtered by the server, so it it is preferable to use the Filter box first and then use the Search box to refine the results. Using the Search box as an alternative to the Filter box could result in missing items.

Synchronization with page URL

The Cluster Manager creates permalinks for all tables that can be filtered or searched. Changes applied using either the Filter or Search boxes are automatically reflected in the page URL, thus allowing you to copy and paste the URL to easily retrieve the same display later. For example, the URL http://localhost:61080/batches/main?status=SUBMITTED would display only batches in a submitted state (assuming the Cluster Manager is running on localhost:61080).

Aborting Batches

Batches can be aborted from the Batches page with the following steps:
  • Click on the ABORT button for the batch you want to abort.
    check jobs
  • Confirm that the selected batch should be aborted.
    confirm jobs abort
  • Aborted batches will show a change of status once the abort has been completed and recorded:
    aborted batches

STANDARD users can only abort batches they launched.
READONLY users cannot abort any batches.
SYSADMIN and ADMIN users can abort batches of any users.

Submitting Batches

The Submit Batch page allows users to submit batches interactively. It provides an alternative to using the grbcluster batch command to submit a batch from the command line (see grbcluster batch --help for details on command-line batch submission).

Submit page structure

Repository

The repository page allows you to upload and manage files that can be reused when submitting a batch.
For example, you may want to upload a single model file but submit multiple batches that solve that same model but using different parameters. You may also want to upload a MIP start (mst) or attribute (attr) file to reuse across multiple models.
repository table structure