CMDB Intelligent Search

1. Introduction

A mature Configuration Management Database (CMDB) in ServiceNow typically contains a huge volume of information related to hardware, software, services, etc. and their relationships. Users frequently struggle to locate the necessary data in their organization’s CMDB. Understanding the structure and schema of the database is usually necessary to query this complicated data model. This is where the natural language query (NLQ)-based CMDB intelligent search capability can really make a difference. It allows users to utilize queries in simple English to retrieve desired information from the CMDB. Thus, they don’t have to come up with any complicated queries or comprehend the underlying database semantics.


2. Accessing and using CMDB Intelligent Search

Step 1: Find CMDB intelligent search from the CMDB Workspace

Click the ‘Workspaces’ menu from the ServiceNow banner and then select ‘CMDB Workspace’ to launch it.

Accessing 'CMDB Workspace'

Intelligent search section can be found at the top of the CMDB Workspace, and a search box is located underneath it.

Finding the search box for CMDB 'Intelligent search'


Step 2: Refer ‘Search tips’ (Optional)

Users can also find suggestions on how to perform the search more effectively. Click the ‘Search tips’ URL located next to the ‘Search’ button to view these suggestions.

Opening 'Search tips'

A search tips pop-up window appears, as shown in the screenshot below.

The following 4 tabs are available on the search tips windows:

  • Single-table search: Tips to search a single CI class and its associated attributes. This is displayed initially when the search tips window opens.
  • Multi-table search: Tips to search multiple classes and related tables.
  • Advanced filtering: Details about the keywords that may be applied as filters to limit the results.
  • Relationships: Tips about how to query relationship data.


Step 3: Search for the required information using natural language

In the search box located under ‘Intelligent search’, users can specify what they require in natural language. E.g., Show me all servers located in San Diego.

Example of a search phrase

Under the search field, we can see past searches and sample searches. As you type, ServiceNow suggests potentially relevant table names and field names.


Step 4: Refine the query if required

ServiceNow will try to convert the natural language into an appropriate database query after clicking search. If there’s still uncertainty, a dialog box will appear, prompting you to clarify certain parts of the query. For instance, the pop-up window shown in the screenshot below lets us enter the name of the table for the CI class that we need to query.

Refining the query

Click the ‘View search results’ button after reviewing the information requested in the pop-up window and providing our inputs if required.


Step 5: Check and interact with the results

The search results will be displayed after clicking the ‘View search results’ button in the previous step. Here we can see options to give feedback on the relevance of the results, export this data, check and modify the filter conditions, etc.

Checking search results

NOTE: If the constructed CMDB query contains more than one table, then the search results will have a ‘View in Query Builder’ button available near it. Click the button to open the CMDB Query Builder with your query fully constructed on the Query Builder canvas. We can use the Query Builder to edit the query if needed.


3. Improving the accuracy of CMDB Intelligent search

The NLQ is pre-trained with a certain level of natural language comprehension. But it is possible to increase its accuracy by configuring the following:

3.1 Synonyms

We can use ‘Synonyms’ to improve the search results by enabling users to find records using different terms whose meanings are the same or similar. For example, if “laptop” is defined as a synonym for “notebook”, searching for either term will yield the same results. This feature helps to make searches more flexible and inclusive. This is configured by navigating to ‘NLQ>Synonyms in the application navigator and defining pairs of equivalent terms. For example, the form view of a record available Out-of-the-box (OOB) on the Synonyms table is shown in the following screenshot. This indicates that a search containing any of the keywords listed in the ‘Synonyms’ field will consider the ‘Name’ of the ‘Model ID’ of the server while fetching the search results.

Configuring 'Synonyms'


3.2 Semantic Shortcuts

Semantic shortcuts enable NLQ systems to understand various kinds of user requests. They map commonly used words to fields in the database. But they are only applicable when certain conditions are met. It is accessed by going to ‘NLQ>Semantic Shortcuts’. Some semantic shortcuts are available OOB. Organizations need to look at defining custom semantic shortcuts by considering various parameters like their requirements, keywords used in past searches (accessible from ‘NLQ>Logs), common use cases, etc.

Accessing NLQ Semantic Shortcuts

In the following example, a search about ‘Servers which I own’ will consider records where the ‘Owned by’ field in the cmdb_ci_server table matches the current user who is doing the search.

Example for NLQ Semantic Shortcut


3.3 CMDB Implicit Relationships

We can improve the capability of Intelligent Search to find relevant results by specifying implicit relationships between CI classes. An implicit relationship defines the relationship between two tables and includes any filters to be applied. Implicit relationships are particularly useful for queries which involve service offerings and application services.

It can be accessed by navigating to ‘NLQ>Cmdb implicit relationships’ from the application navigator. The relationship tab of the ‘Search tips’ window also has a link to access this.

Accessing 'Cmdb implicit relationships'


4. Conclusion

The CMDB intelligent search enables ServiceNow users to gather the required information from the CMDB without having to comprehend the database complexity. As a result of improving the user experience while querying the CMDB, it may indirectly increase the effectiveness of CMDB in a variety of use cases, such as service operations, compliance management, IT asset management, planning, etc. Intelligent search enables more stakeholders to tap into the authoritative CMDB intelligence through natural conversation.

Date Posted:

June 26, 2024

Share This:

Leave A Comment




Fresh Content
Direct to Your Inbox

Just add your email and hit subscribe to stay informed.