Scheduling semantic model processes

Applies to: Kyvos Enterprise Kyvos Cloud (SaaS on AWS) Kyvos AWS Marketplace

Kyvos Azure Marketplace Kyvos GCP Marketplace Kyvos Single Node Installation (Kyvos SNI)

You can set up the schedule for all of the process types. Schedule the frequency and the time when semantic models get processed. You can also set semantic model process properties. To learn about reverting to or managing backup processes, see Managing semantic model job instances.

Note

From Kyvos 2023.3 onwards, you can cancel Drop partition jobs.
From Kyvos Manager, you can set the duration (in hours) to limit the semantic model process time using the BUILD_TIME_THRESHOLD (OlapEngine property). Once the semantic model process limit is reached, the process will fail. If you resume the semantic model process, it will use the same duration that you set during the initial semantic model process.

There are two options for scheduling the different process types. See Semantic model process types. Click the i icon next to the job type list to get a description of each process type.

Schedule process
Add Other Jobs

Other job types

To schedule other job types, click the ... menu on the Process tab and choose Add other jobs option.

For Data Profile processes, set up sampling options, the schedule, and column profile. See Creating a data profile.
To Drop Partitions, you can add or remove rules and add filter criteria. See Drop partition from semantic model.
For Cache Processes, set up cache rules. See Setting up and using cache rules.

Frequency

You can specify the frequency of when your semantic model gets updated. For example:

Every X HOURS
Every X Minutes
DAILY - at a given Time (HH:mm)
WEEKLY - on a given weekday (Sun-Sat) at a given Time (HH:mm)
MONTHLY - on a given day (1-31, First, Last) at a given Time (HH:mm) on a given day (First, Last weekday) at a given Time (HH:mm)

When you choose Full or Full Test data process, you can specify to process the source databases if modified (if changes have occurred) or process always. Depending on the job type you select, you can also define whether to process once per day or repeat. You can also view the properties currently in use or change them.

Resuming after process failures

If you resume a failed process, the steps that were successfully completed are skipped reducing the process time. There are several ways to resume a process after it fails to complete.

From the Process tab of a semantic model, right-click a failed job and choose Resume job. This option is available when the process failed after some of the steps were successfully completed.
When you add a job and have selected a semantic model that failed to process on the previous attempt, you may see an option to resume from the last failure. You are prompted to confirm that you want to resume the process.

Modifying semantic model design after process

Typically, to modify the semantic model design after a process, you need to perform a reprocess (incremental or full prcoess) for the changes to come into effect. However, you can modify certain entities and the partition and aggregation strategy of a semantic model without the need for reprocessing. See Semantic model modifications allowed without reprocess to know more.

For this, perform the following steps.

Navigate to the semantic model that you want to modify.
Make the necessary changes on the designer screen and click Save.

Related topics