Alma Jobs

Last updated
Save as PDF
Share
1. Share
2. Tweet
3. Share

Introduction

Alma jobs allow running batch processes on a large number of entities. The jobs run in the background, without impacting the day-to-day work of the library staff. To maximize efficiency, the system constantly performs optimization and prioritization of the running jobs in order to make sure that each job gets enough resources to complete. In addition, some jobs can run in parallel (Multi-threaded) – the job population is split into parts that can run in parallel to each other. Before a job is executed, it is examined by the system to check whether any dependent jobs or other instances of the same job are currently running.

For information, watch Applying Changes to Batches of Records in Alma (Jobs) (06:00).

Job Types

Jobs are sometimes called processes in the UI.
Some Alma jobs are dependent on other jobs due to the fact that they modify the same data. A job can also modify data that has a relationship with data updated in other jobs. To prevent these jobs from failing due to simultaneous data use, dependent jobs are scheduled to not run at the same time.
The following dependent jobs wait till the completion of the job they are depended on with the status 'Pending', and start running after that job completion: Authorities - Link BIB Headings, Authorities - Preferred Term Correction, Authorities - Handle Local Authority Record Updates, MMS - Build Record Relations, General Publishing.

Scheduled Jobs

Scheduled jobs run periodically. Some of these jobs are scheduled by Alma. For other jobs, you can configure the schedule using predefined job scheduling options. Note that scheduled jobs run as close as possible to their scheduled time. A job may wait until other jobs are done and the system has enough available resources.
In addition to their scheduled run times, Alma provides the ability to run some of these jobs at any time. For example, you can run an import job on demand (see Running Import Profile).

Workflow Jobs

Workflow jobs run automatically when they are required. For example, after submitting a purchase order, the job Export Orders (PO) Job runs to send the PO to the vendor. In some cases, you can also run or rerun these jobs at any time—for example, if the original run failed.

Manual Jobs

These jobs are available to run by you as required. Some manual jobs first require you to create sets (of items, users, and so forth) for the job to process.
To run manual jobs, see Manual Jobs.

Configuring Jobs by Use of Profiles

Alma uses profiles to configure and/or schedule certain types of jobs. You add the profile, give it a name, and configure its parameters. This creates a job that is available to run, either from a specific location within Alma or automatically as a scheduled job.

Common profiles in Alma include:

Integration Profile – Creates a job that exports information to, or imports information from, an external system, such as a Student Information System. For more information, see Configuring Integration Profiles.
Import Profile – Creates a job that imports bibliographic or authority records into Alma, and may also import purchase order or inventory information. For more information, see Record Import and Monitoring Import Jobs.
Publishing Profile – Creates a job that exports bibliographic records from Alma to an external system, such as Primo or Google Scholar. For more information, see Configuring Publishing Profiles.

Jobs Monitoring

To view scheduled jobs, see Viewing Scheduled Jobs.
To view running jobs, see Viewing Running Jobs.
To view job history and reports, see Viewing Completed Jobs.

The option to Abort a run job is further described in Viewing Running Jobs.

Network Zone Perspective

Each job running by an institution can be monitored from the Network Zone. When viewing the jobs from the Network Zone, an option to filter on a specific institution is available. When this option is selected, only jobs of the specific institution will be visible. The Network Zone cannot execute a job on behalf of an institution.

Network Zone - Job monitoring.png

Job Dependencies

Some jobs have dependencies with related jobs (for example, the Bibliographic Heading Correction job cannot run when the Linking job is running) or with other instances of the same job (for example, the Loan Anonymization job), which means they cannot run at the same time. The job will be finished with a message that it cannot be executed when the other job is in process. For jobs that are scheduled by the system, the time scheduling of these jobs is set up to minimize overlapping.

Expected Job Duration

As a multi-tenant cloud-based solution, Alma has a sophisticated batch job management architecture, which takes into account various factors such as the types of jobs running, the time of the day, and the general load of the system, all in order to provide you with the services you require.

When using Alma’s batch job services for managing your repository, it is useful to know how long processes can be expected to run. The guidelines provided below are based on Alma’s actual production use of these services. Ex Libris expects that, on average, the job capacity will not be less than listed, and often even better. This depends on the system load at any given time.

Batch Job Planning Guidelines
Batch Service	Records Per Hour (Peak)	Records PerHour (Off-Peak)
Metadata Import Note: The different profiles of metadata import may influence the time the service takes. For example, an EOD import profile will on average take longer than a bibliographic only update.	30k	120k
Metadata Export	150k	600k
Global Changes – MARC Bibliographic Normalization	50k	200k

Off-Peak hours are during the night (relative to your data center location) and usually spans 6 hours (for example, midnight to 6:00 AM).

For more information, see Scalability and Performance.