Configuring cache recommendations in Data Virtualization

Data Virtualization Managers can configure cache recommendations to improve query performance.

About this task

Data Virtualization can analyze your queries to generate cache recommendations, which are based on the following configuration settings that are specified by the Data Virtualization Manager.
  • Queries that were run in the past n days.
  • Minimum average execution time of queries (minutes).
  • Minimum number of times queries were run.
  • Number of recommendations to fetch.
  • Whether to enable machine learning based recommendations. By default, machine learning based recommendations are enabled.

In the Recommended data caches tab of the Cache management page, as Data Virtualization Manager, you can configure cache recommendations to improve query performance.

Procedure

  1. Go to Data > Data virtualization > Virtualization > Cache management.
  2. Click the Settings icon Settings icon on the Recommended caches tab.
  3. Enable or disable machine-learning based recommendations.

    When you toggle machine-learning based recommendations and click Save, the ranking and number of recommendations might change. If you enable machine-learning recommendations, changing the time period of queries can also change the recommendations.

  4. Specify the number of cache recommendations to show in tables. Select All or Specify a number.

    If you select Specify a number, you can use the slider to specify that 1 to 10 recommendations appear in the table. The default is All.

  5. Use the slider to specify the query time period, in number of days, 1 - 15.

    The default is 1.

  6. Specify the minimum average query execution time (in minutes) to consider in cache recommendations.

    The default is 1.

  7. Specify the frequency to consider for queries that are repeated often.

    For example, if you specify 5, only queries that were run at least five times are considered, and if you specify 1, every query is considered.

    The default is 1.

  8. Click Save.
  9. Click Find recommendations to trigger the recommendation engine using the current settings in Data Virtualization.