Configure general search behavior
You can customize the search index configuration.
Prerequisites
You have the ADMIN or SUPER role in Collibra Console.
Steps
- Open the DGC service settings for editing:
- Open Collibra Console.
Collibra Console opens with the Infrastructure page. - In the tab pane, expand an environment to show its services.
- In the tab pane, click the Data Governance Center service of that environment.
- Click Configuration.
- Click Edit configuration.
- Open Collibra Console.
- In the Search Index configuration section, enter the required information.
Setting Description UI search appends wildcard *
True (default): A wildcard (asterisk) is automatically added to the end of each search text to find more matches. For example, search text loan sizeis treated asloan size*, finding assets containing sizeable.An asterisk isn't added if:
- The search text contains a tilde (~).
- The search text ends with a quotation mark (").
Note This setting applies only to search texts via the user interface. A wildcard isn't automatically added to search texts via REST API.
False: A wildcard isn't added to the search text.Note In existing environments, the current status of the setting is retained. This means that if the setting was previously disabled, it will remain disabled.Maximum batch size * The maximum batch size for resource reindex.
- Default value: 5,000
- Minimum value: 100
- Maximum value: 30,000
Maximum batch size for relations *
The maximum batch size for relation reindex.
- Default value: 500
- Minimum value: 10
- Maximum value: 10,000
Maximum batch size for relation path preview *
The maximum batch size for the relation path preview of relation index.
- Default value: 500
- Minimum value: 50
- Maximum value: 10,000
Stop words (retiring)
(Requires restart)
This setting requires the SUPER role.
A list of stop words that are ignored as tokens for the index.
The default list of English stop words includes the following:
a, an, and, are, as, at, be, but, by, for, if, in, into, is, it, no, not, of, on, or, such, that, the, their, then, there, these, they, this, to, was, will, with
If you don't create your own list of stop words, the default list is used.
If you create your own list of stop words, you need to:
- Reindex Collibra.
- Restart the environment to apply your changes. For more information, go to Stop an environment and Start an environment.
Relation-based search
True (default in new environments): The Data Marketplace search considers certain assets and relation types between assets. As a result, your search results not only include assets that directly match the search criteria, but also assets that match the criteria through specific relation types.Example A column named Order is included in a data set named Customer. If the relation-based search is enabled and you search for Order in Data Marketplace, then the data set Customer appears in the search results because the data set contains this column.
Tip For more information about this feature and the default relation types, go to Filtering and searching based on relations in Data Marketplace.
False: The Data Marketplace search results do not consider relations.After you enable this setting, you must reindex Data Marketplace relations or reindex Collibra completely.
Note In new Collibra environments, this setting is enabled by default. In upgraded Collibra environments, the previous status of this setting is retained.
Number of threads for multi-threaded resources indexing *
This setting requires the SUPER role.
- Determines how many threads are used for multi-threaded resource reindexing. You can enter a value between 0 and 32. If you enter 0, the number of threads is automatically calculated based on the number of cores. This setting is applied only when the Multi-threaded resources indexing setting is enabled. If the entered value exceeds the number of available CPUs, the number of threads defaults to the number of CPUs. For example, if you enter 32 threads but only 4 CPUs are available, 4 threads are used for indexing.
Number of threads for multi-threaded relations indexing *
This setting requires the SUPER role.
Determines how many threads are used for multi-threaded relations reindexing. You can enter a value between 0 and 32. If you enter 0, the number of threads is automatically calculated based on the number of cores. This setting is applied only when the Multi-threaded relations indexing setting is enabled. If the entered value exceeds the number of available CPUs, the number of threads defaults to the number of CPUs. For example, if you enter 32 threads but only 4 CPUs are available, 4 threads are used for indexing.
Use managed Elastic
This setting requires the SUPER role.
True: Managed Elastic deployment is used to handle indexing.
False (default): Managed Elastic deployment isn't used to handle indexing.Note- This setting is applicable only to commercial customers, except Azure.
- If you want to enable the setting, contact Collibra Support.
Auto full reindex on incremental threshold exceeded
This setting requires the SUPER role.
When enabled, Collibra automatically triggers a full reindex when incremental indexing events exceed the threshold. This setting applies only when managed Elastic is enabled.
Threshold for incremental indexing events to trigger full reindex automatically *
This setting requires the SUPER role.
When Auto full reindex on incremental threshold exceeded is enabled, this setting defines the minimum number of incremental events required to trigger a full reindex automatically.
The threshold range is between 1 million and 100 million.
This setting applies only when managed Elastic is enabled.
Quick Search - Expanded Scope
This setting requires the SUPER role.
Expands Quick Search beyond names to include descriptions and definitions, improving relevance and surfacing better-governed content. If no name-based matches are found, Collibra suggests relevant results from descriptions or definitions.
- Click Save all.