
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
04-25-2023 04:35 PM
Hello SNOW CSDM Team,
Has anyone encountered the need to track availability for individual application services using a service offering? If so, is there a diagram that illustrates how this fits into the CSDM 4.0 model? It seems like a reasonable use case to use a service offering to track availability for an application service when needed (example: to tie back actual availability to a vendor contract). Just wondering if this was left out of the visual models for a reason.
I appreciate your thoughts and any documentation for reference.
Solved! Go to Solution.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-03-2023 05:29 AM
I'd like to understand why ITOM operational services (Application Services) are frequently included in CSDM discussions about SPM Services. ITOM and SPM "Services" live in 2 different hierarchies in the CMDB, representing 2 totally different concepts (despite the fact that they unfortunately share the same base class). If you look carefully at the CMDB Schema, ITOM Application Services and SPM catalog Services do not interact directly with each other(see details below).
ITOM Services
- ITOM Application Services (labeled in the CMDB as "Business Services" prior to CSDM) model sets of deployed Business Applications that are deployed to Infrastructure Services (e.g., farms of Web, Application and Database servers) consisting of Network and Server CIs.
- Infrastructure Services are modeled using Dynamic CI Groups (labeled in the CMDB as "Technical Services" prior to CSDM) which are a subclass of Application Service.
- ServiceNow ITOM applications provide the tools(e.g., Operator Workspace) needed to manage the availability of the operational service hierarchy from the highest level Application Service Group down to the most detailed CIs created by discovery to model hardware and application components and configuration.
- The availability and state of infrastructure CIs used by Application Services directly affect the operational state of the Application Service, so they are monitored closely. Alerts propagate up to the top of the operational hierarchy. The current operational state of all ITOM Services can be viewed using the operator workspace dashboard(which is usually monitored 24x7 by most companies).
- Each Applications Service Group contains Application Services and Application Service Groups it operationally depends on. The highest level Application Service Groups represent the operational state of Corporate-level Business capabilities (e.g., Lines of Business like Manufacturing, IT, Sales, Support,Marketing) which are based on operational state of the Application Services the depend on. As you move down the hierarchy, you will eventually see Service Maps for each Application Service.
SPM Services(including Service Offerings)
- SPM Services were created to model the taxonomy of requests made available to consumers through Service Catalogs/Portals (They are all part of the CSDM Sell/Consume Quadrant)
- SPM Services spawn Service Requests that are defined in and execute in the ServiceNow Cloud, not in the ServiceNow customer's Data Centers.
- Service Requests may trigger automations(e.g., Server Build Scripts) that run in the customer managed Data Centers. These automations could be modeled in the CMDB as batch jobs that are part of an ITOM Application Service(Dynamic CI Group) that has an operational state. Each of these batch jobs may have a relationship to the CIs they create, but the SPM Service(which is a template for a Service Request) doesn't have relationships defined in the CMDB to operational CIs (e.g., Hardware and Application CIs).
- If required, critical SPM Services and Service Offerings URLs could be modeled as manually created CIs belonging to a 3rd Party(SAAS) Application Service called "ServiceNow Catalog Service" (Represented as a Dynamic CI Group because SAAS entry points can't be discovered by Service Mapping).
- To summarize, SPM catalog Services and Service Offerings have no operational status, even if Service Requests they spawn were represented as operational CIs(e.g., Batch Jobs).
Is an ITOM Application Service a service?
- Can we all accept Microsoft's use of the term "Service" (as used in Microsoft Windows when application functions are bound together to "run as a service")?
- For example, MID Server Applications bundled together and deployed on Windows to "Run as a service".
- Microsoft Services execute, can be monitored, and have an operational status just like Application Services defined by ServiceNow.
The answer is Yes.
Is an SPM Service a Service?
- They don't "Run" on customer infrastructure or on the ServiceNow cloud. They are more like metadata for Service Requests that "Run" in ServiceNow.
- They are not Monitored
- They don't have an operational status
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-03-2023 05:46 AM
I disagree that using Microsoft's definition of a Service is the gold standard. If you look at Linux, what Microsoft defines as a Service, Linux defines as a Process. ServiceNow is using the term Service as ITIL defines it, which is closer to the layperson's definition. Service is delivering valuable outcomes to consumers by performing work that you have the skills, tools, time, and willingness to perform, assuming costs and risk. Like getting your car services. Like ordering room service. Like a home inspection service. Like food service. An IT service is simply any service that involves information technology. When we do IT Service Management, we are focused on service levels that we commit to with the customer. Microsoft is the odd-ball here. It's more of a stretch to think about some obscure computing process as a "service". So no, Application Service is not as clear an example of a Service, although it's a bit closer than a running process, because arguably there is some more tangible outcome involved in running the entire stack which is actually accessed by the users. Still, I think better terms would be Instance or System, and that seems to be the direction they are heading. I think the main reason for Application Services extending the Service class is so that the Impacted Services would pick it up and provide a basic actionable business context for IT Service Management processes for customers who have not yet abstracted their application usage up to the true Service and Offering level.
The opinions expressed here are the opinions of the author, and are not endorsed by ServiceNow or any other employer, company, or entity.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
05-03-2023 07:58 AM
Thank you for your thoughtful response. I think there is a third use case for using the Service Offering function to monitor the availability of what you have referred to as an ITOM Service. The current CSDM model doesn't appear to accommodate for this unless the ITOM service is reclassified as a Technical Service or a Business Service. I appreciate your help.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
04-26-2023 12:03 AM
I think I remember, that the Platform, introduced more availability calculation on various levels, not only the offerings, has always been a challenge, here is link
View availability results (servicenow.com)
But that said, if it is tied to a contract, then use either business service offering or technical service offering and link these to application services, as said by Barry.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
04-26-2023 06:09 AM
While logically it makes a ton of sense to calculate availability on the application service itself, I have long accepted that the reason for the current architecture is specifically that the Service Offering, by definition, is where we calculate service performance metrics. And accepting this universally actually will make your life simpler, since you know that by definition, service level management activities are always conducted at the Offering level. Thus, in order to track availability for a specific Application Service, there should be a 1-to-1 relationship with an appropriate Offering that is specific to that Application Service. This fits well in CSDM and is logically justifiable. If there is a reason to track an Application Service's individual Service Availability, that is all you need to justify giving it a specific Service Offering as well. That may seem like duplication, but we need to think of these different layers from more of a "composition" approach, in that each CI represents a different aspect of the overall picture. The Application Service represents the group of infrastructure CIs and their relationships to one another, while the Service Offering represents how we are delivering outcomes to consumers and how we are measuring service levels and commitments.
The opinions expressed here are the opinions of the author, and are not endorsed by ServiceNow or any other employer, company, or entity.

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
04-26-2023 08:30 AM
Thank you!