Improving Hardware and Consumable Product Model Normalization
Use Case
You are the Hardware Asset Manager for a ServiceNow HAM customer. You have noted not all your Hardware Product Models are fully normalized and you might be questioning “why?”, or “is there something wrong with our product configuration?” While there might be something wrong with your configuration, it is likely something simpler, such as not being opted-in to a HAM Resource Category or missing data in an evaluated attribute.
Proper normalization can set the stage for successful hardware asset management by helping to remove a lot of the “guesswork” out of the work we do. Let’s take a look at how Hardware Normalization works and how to understand your normalization results.
Note: For this blog, I’ll address Hardware Product Models. The same processes and concepts apply to Consumable Product Models with the exception that the business rule on the cmdb_consumable_product_model table is Normalize consumable model not Normalize hardware model as identified in the next section.
How Hardware Model Normalization works
On the Fly Normalization
Hardware (HW) model normalization utilizes an “after” Business Rule entitled Normalize hardware model which executes against records inserted into, or updated on, the cmdb_hardware_product_model table. Filter Conditions within the business rule allow execution when certain attributes on the HW model are changed, the HW model isn’t already manually normalized, and when the License Opt-In status matches the HAM Resource Category. You can opt-in or out of a HAM Resource Category through the HAM Resource Categories module. In the screenshot below, none of the HAM Resource Categories are opted-in, which is adversely impacting the normalization rate of HW Models in the ServiceNow instance.
When the Normalize hardware model Business Rule executes against a record, values stored in the Name, Manufacturer, and Model Number attributes from the record header and General tab are passed to the Hardware Model Normalization Engine API for evaluation. If matches are found and rules are present, the engine updates the attributes located on the Normalization tab
Values utilized by the Hardware Normalization Engine
Output from the Hardware Normalization Engine
Scheduled Normalization
Normalization is also executed daily by a scheduled job. Weekly content downloads are randomized so the normalization job is executed daily to ensure the instance always accounts for the most recent data available from ServiceNow’s Content service. The job will also account for models whose Resource Category may have recently changed from opted out to opted in.
The name of the daily scheduled job is HAM – Hardware Model Normalization. The scheduled job evaluates records in the same fashion as the on-the-fly method identified above, however, it does so for all model records in the table that are not fully normalized, not manually normalized, and are not in an opted-out Resource Category.
Understanding Normalization Status Values
The Normalization Status can be one of the following values:
- New
- The record has not been evaluated by the Normalization engine.
- The Normalization tab will be empty with the exception of the Normalization status attribute, which is set to “New”.
- Normalized
- The record was found and fully normalized
- The Normalization tab will have values set for Normalized manufacturer, Product, Model, and Device type.
- When present in the content library, Hardware Model Lifecycles will also be written to the Hardware Model Lifecycles related list
- Manufacturer Normalized
- The hardware manufacturer was identified by the Normalization engine while the Product and Model were either not matched or normalization rules were not present to complete the normalization.
- Partially Normalized
- The hardware manufacturer and the product are identified but the model is not matched. This status can also occur when the manufacturer, product, and model are identified but a normalization rule is not present that matches the model to the product.
- Match Not Found
- The record was processed by the normalization engine and none of the supplied data was recognized.
- Manually Normalized
- One or more values on the Normalization tab were manually entered by a user.
Achieving Better Normalization Results
There are three primary factors that drive normalization accuracy; opting-in to resource categories, quality of the data sent to the normalization engine, and the depth of the data contained in the content library. To achieve a high rate of fully normalized hardware model records, you will want to be sure your hardware model data is as accurate as possible.
Opt-in to Hardware Model Resource Categories
HW models will not be normalized if they are not part of an opted-in Resource Category. Starting in the ServiceNow Rome family release, the ability to manage your HAM subscription cost through active management of HAM Resource Categories was introduced. If a certain category of hardware models are all reflecting “New” as their normalization status, then you are likely opted-out of that category. This is the state of the demo instance from which I presented a screenshot above, where none of the Resource Categories reflected a “true” Opt-In status.
Before opting-in to a Resource Category:
Be sure you understand the impact on your ServiceNow subscription costs and
that you have executive leadership approval.Once you are opted-in, normalization is initiated for all records in the chosen HAM Resource Category.
Improve Your Existing Data
HW Models can be created either manually or via automation. When entering data manually, be certain to accurately enter the Manufacturer, Name, and Model number into their respective attributes in the record header and the General tab. When HW model records are created via automation as part of a discovery source integration, complete data may not always be returned for a scanned device, or data critical to full normalization may be concatenated into the “Short description”. A lack of complete data will usually result in a Normalization status value other than “Normalized”.
In the following screenshots, we can see that the Model number attribute is blank and the Short Description appears to have the model number appended after the word “router”. For this record, the normalization status is “Partially Normalized”.
With a bit of user intervention, you can improve this normalization result by copying the model number from the end of the “Short description” and populating it into its real home, the model number attribute. Once done, save the record and the Normalize hardware model business rule will execute taking into account the model number. Caution! Do not populate the model number attribute on the Normalization tab with this data. Doing so will result in a “Manually Normalized” normalization status. Be sure to enter the value into the Model number attribute on the General tab.
Set up the Hardware Asset Content Service
Configuring the Content Service on your instance will send an anonymized feed of not fully normalized HW models to be reviewed by the ServiceNow Content Curation organization. The team is responsible for researching and adding new HW model content data to the Content Service, which is then pulled into customer instances by weekly scheduled jobs. The team includes hardware data from many sources but one of the most valuable is the anonymous feed from customer instances because that data reflects real-world roadblocks to ITAM success. Product documentation on configuring the Content Service in your instance is located here: Opt-in to the Hardware Asset Management Content Service.
Additional Self Help
While the data sent to the Content Service is being reviewed, you may not have the luxury of waiting for your Hardware Models to make their way through the curation and QE cycles. If this is the case, there are additional features available to assist you. Use of any of these features will result in a Normalization status of “Manually Normalized”. This status isn’t a bad thing, it just means that someone within your organization completed the normalization process, and identification wasn’t completed via the ServiceNow normalization engine. If you utilize this method, you should consider researching manufacturer lifecycle dates and adding them to the Hardware Model Lifecycles related list.
Custom Products
Custom Product records can be created for any product not found in the Content library. Once you’ve created a Custom product you can use that product on the Normalization tab to manually normalize a HW model. Documentation for Custom Products is found here.
Custom Models
Custom Model records can be created for any model not found in the Content library. Custom Models can be related to products found in the Content Library or a Custom Product. Documentation for Custom Models is found here.
Complete Normalization Manually
You can also complete normalization manually. In the screenshots below I will show you how to complete normalization where a normalization rule (part of the content library) either isn’t present in the system or is not making a match. In this scenario a Cisco Systems, 4999 Integrated Services Router is Partially Normalized. All the data is in the correct attributes in the header and on the General tab, however, the model isn’t being matched by the normalization engine resulting in a “Partially Normalized” Normalization status.
To complete the match and finalize normalization, look up the model number using the “magnifying glass” on the Normalization tab’s Model attribute. The models available are a combination of models supplied by the Content Service and those created as Custom Models by an end-user.
Once you've selected the Model, save the record. The result is a Normalization status of "Manually Normalized" as reflected in the screenshot above.
Learn More About ServiceNow ITAM Products
To learn more about ServiceNow ITAM products, visit Now Learning and choose IT Asset Management from the “Select Product” section in the left navigation. You will be presented with both paid and free resources featuring a number of on-demand webinars, classes both on-demand, and instructor lead, as well as full learning paths designed to prepare you for product certification.
2 Comments
