Self-Service Analytics queries each connector to better understand its data store’s capabilities and behavior. The connector’s response describes to Self-Service Analytics the Self-Service Analytics features that the connector and the data store can support and any limitations to that support. It identifies the type of data requests that the connector and its data store can fulfill.
To learn more about a Self-Service Analytics feature and the connectors that support it, select the feature from the list below, organized by feature category:
Advanced Visualizations
Group By Functionality
Filters
Metrics
Security
Fused data sources inherit the limitations of the underlying connectors used by the fused sources. In addition, fused sources have other feature limitations. See Data Fusion Limitations. Custom SQL Queries
Applicable only to SQL-based connectors, a data source using a connector that supports custom SQL queries can use an SQL query to select fields from the table. The custom SQL statement can be specified on the Custom SQL area of the Source Creation tab after selecting the Custom SQL option. Any visual you create displays fields in the order they are retrieved from the source. When you create a source using custom SQL, your field data is shown in the order you specify.
Custom SQL queries are a powerful tool for performing complex data queries. However, be careful when creating custom SQL queries because it is easy to define a heavy query or a query that may overwhelm your database. Use this feature carefully.
Fast Distinct Values
Connectors that support fast distinct values can efficiently return distinct values for a field. This functionality optimizes the retrieval of distinct (unique) values in large numbers of records. If a connector supports this feature, the Filter dialog is populated with distinct values for an attribute directly from the data source, without the need to refresh the data and without retrieving or storing the distinct values in the metadata. For example, Elasticsearch keeps lists of distinct values at the ready. Features such as these make fast distinct values possible for your connector.
There is no metric that defines “fast”. This value is based on the judgment of the developer.
When custom ranges or list values are requested for a field, full data scans are not performed.
For most connectors, this feature can be safely left disabled without impact.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Group By Multiple Fields
Many connectors can group by more than one field in a query. Here is a sample SQL query:
If multi-group querying is not supported, some visuals will be unavailable for a data source.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Group By Time
Many connectors support grouping data by a real time field. This function is a prerequisite for all time-based visuals.
Most commonly, the data includes a date or time stamp field type that corresponds to Self-Service Analytics’s date field type. Here is a sample SQL query:
If grouping on time is not supported, some visuals will be unavailable for the data source.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Group By UNIX Time
Many connectors support grouping data by an integer field that contains times in Unix (epoch) time.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Histogram Floating Point Values
Many connectors support the calculations necessary for histogram visuals with non-integer values, such as floating point (32-bit) and double-precision (64-bit) floating point data types.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Last Value
The last value metric in data is the last value in all the data values for a field, sorted by the time attribute selected for the time bar. If the latest date and time for the time attribute is exactly the same in multiple records, the last value for the field is the maximum value of the field in the records with the latest date and time.
When a connector supports the last value feature, it determines and uses the last value of a selected field in the data.
Although many data stores implement a last value function, the Self-Service Analytics last value indicates that the last value in a given field collection can be loaded and used in a visual immediately.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Multivalued Fields
Some connectors support aggregation by multivalued fields, such as arrays.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Partitions
Some connectors support partitions and pruning. This feature enables the Partition column on the Fields tab of the data source configuration, when the data source uses a supporting connector. Partitioning allows you to link a partitioned field to another field to help improve the performance of filtering operations for a data source.
Although many data stores support partitions in some form, this feature specifically tells Self-Service Analytics that the partitions may be used for manual pruning of result sets to increase speed.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Schemas
When a connector supports schemas, it supports namespace, schema, or catalog notation for organizing collections. When schemas are supported, the Source Creation tab of the data source configuration displays a Schema drop-down you can use to select a schema for the data source configuration. Elasticsearch has a custom UI for displaying multiple indices.
Visualize the relationships in your schemas in supported data sources, and connections. Add more relationships in connections as needed. See Visualize Schemas and Joins
If you would like to make a default schema available to your users, or hide a schema from your users, update the properties file of your connector. See Select Schemas .
Select Schemas
Control which data source schemas are treated as internal by Self-Service Analytics and are not disclosed to users during source creation. Supported schemas are included in the Schema drop-down selector in the Source Creation tab of the data source configuration.
Add or edit the system.schemas property to the properties file of supported connectors to specify the schemas to use. Restart the connector after making these changes. The schemas you want to make available to users are visible in the Schema drop-down.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Text Search
When a connector supports text searches, it can perform an efficient search on text fields. When text searches are supported, search control is enabled on dashboards using data sources that use the connector. To turn on text search, Enable Text Search in the Global Settings tab for your sources.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
TLS
Many connectors support SSL/TLS encryption.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Wildcard Case-Insensitive Filters
Many connectors support case-insensitive wildcard filters.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
Wildcard Case-Sensitive Filters
Many connectors support case-sensitive wildcard filters.
Support for this feature by connector is shown in the following table.
Key:Y - Supported; N - Not Supported; N/A - not applicable
