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 cache rules to improve the performance of queries.

From Kyvos 2025.3 onwards, an elastic search service is supported.

From Kyvos 2025.2 onwards, auto cache support for SQL generated through Kyvos Dialogs is available.

To set up cache rules for a single semantic model, perform the following steps.

1. From the Toolbox, click Semantic Models.

2. Select a semantic model or create a new one.

3. Click the Refine tab.

4. Click the Auto Cache Strategy tab.

5. For Cache strategy on every semantic model process and Cache strategy on every cache process job, choose cache strategy when it's reprocessed from:

   1. Purge and Populate Cache

   2. Purge Cache

6. You can now set cache rules.

7. Select the types of queries to cache from:

   1. Metadata Queries: Select this when raw data querying is enabled, and the semantic model or its dimensions are not materialized. By default, metadata will be populated for all non-materialized columns. To populate metadata for specific columns, click the Choose columns link and select the columns.

   2. Historical Queries: Select to generate and cache linked filter queries, and choose one of the options from Auto, Yes, or No.
      By default, Auto is selected.  

8.  To set cache population rules for historical queries, do the following:

   1. Select the field, criteria, and value:

      1. For Query Type: Select MDX, SQL, Kyvos Worksheet, Kyvos Chat Session, or Select All, and specify whether they are in or not in the list. By default, the Worksheet comes under MDX.

      2. For Query Status: Select the status from Success, Failure, or Canceled, and specify whether they are in or not in the list.

      3. For Query Completion time: Specify whether it is in the last hour, today, or set the value, such as the number of days/months/weeks/hours/minutes.

      4. For Query Elapsed time: Specify whether the value is greater than, less than, in between, or equal to the value you select in hours, minutes, or seconds.

      5. For User: Specify whether the account you seek starts with, is, contains, or ends with the value you specify.

         Open

         

   2. Use the parenthesis with And or Or to develop more complex criteria.

   3. Click the Plus icon to add more rules or the Minus icon to remove a rule. 

9. Click Save.

Open

You can also specify a cache population job schedule for specific semantic models.

To apply a cache rule to specific semantic models, perform the following steps.

1. Click a semantic model from the list and click the Process tab.

2. Click the Actions menu (...) menu and choose Add Other Jobs. The Add Job dialog is displayed.

3. Select the Job Type as Cache Population.

4. Add tags to label the job instance.

5. Click the required schedule and specify when it should run.

6. Click Schedule.

   Open

   

7. Click Schedule to schedule or start the process.

Enabling Elastic Search service

To enable an elastic search service, you must select the Elastic Search Value Indexes checkbox and then run the Cache population job.

To enable elastic search, perform the following:

1. From the Toolbox, click Semantic Models.

2. Select a semantic model or create a new one.

3. Click the Refine tab.

4. Click the Auto Cache Strategy tab.

5. Click the Elastic Search Value Indexes checkbox and then specify the number for the Maximum Cardinality value. Search values will be created for all the attributes, levels having distinct values below the configured value.

6. Select the Additional Columns from the list for which search entries will be unconditionally created even if their cardinality is above maximum cardinality threshold.

Open

Related topics

• Improving query performance using caches

• Job history

• Sliding window