- Post History
- Subscribe to RSS Feed
- Mark as New
- Mark as Read
- Bookmark
- Subscribe
- Printer Friendly Page
- Report Inappropriate Content
3 weeks ago - edited 3 weeks ago
Technical Services versus Application Services
One of the most common misconceptions I see in the CSDM is folks wondering the difference between a Technical (or Technology Management, now) Service - and an Application Service (or Service Instance, now). We'll use the new terms hereon in.
For the purpose of this article, assume we're talking about Active Directory, Entra ID or other "very internal, not at all user-facing, but way critical to the Business" type things, because that is what I frequently hear as having been mapped as a Technical Service - which isn't quite right.
Why the challenge?
The CSDM does a great job of enabling you to articulate what Business dependencies exist - but it doesn't explain very well that not all Services your Organisation maintains will have one. If you look at the below diagram you very likely assume that each Service Instance must provide for a Business Service Offering (and frankly, that's fair enough. More on that later).
So what do we do instead?
Firstly, the CSDM is so much easier when you start with what you know. Let's start with the below:
1. A Technology Management Service (or Technical Service, previously) articulates who supports according to which SLA or activity or Location.
2. A Service Instance (or Application Service, previously) is a deployed instance of something, for example Production, Pre-Production - and now with the rename to "Instance" we're clearer that it's not Application centric. Thank goodness for that.
Logically, then, we can't state that Active Directory is a Technology Management Service - because it's not articulating who supports it - but that it exists. Which means, then, that (like we're filling in a crossword) Active Directory Production doesn't fit in Technology Management Service, but into Service Instance.
But what about the Business Service Offering?
Remember earlier I said "more on that later"? Here we are. I'll come right out with it:
Not every Service Instance supports the Business Directly.
Before you throw your Business Services on a fire and start dancing around it, we'd better caveat that. Effectively, some platforms and tools - such as Active Directory, ServiceNow, Microsoft Azure - don't directly provide value to the Business. It's what you deliver on them which does. Therefore, for those type of things - based on our previous understanding of what a Technical Service is - we create Service Instances, but we instead link them to other Service Instances which do create value for the Business.
Imagine your ServiceNow logins are all controlled by Active Directory, and it might look something like this:
Notice from the above that each Service Instance has its' own Technology Management Offering - a group that supports it - but not all directly articulate Business Value. In fact, in this example, only Incident Management does.
Hopefully you see that by modelling what some call "Technical Services" as Service Instances, we are able to better articulate the impact of Changes, Alerts and understand the dependencies from those Service Instances which do directly support the Business. Exactly what that all looks like in practice though, will be for another time.
As ever, if you've any questions or feedback, please post them below. I'm always happy to hear perspectives, be told (constructively) why I'm wrong and so on. If this helps you in any way, I'm glad.
#CSDM #ServiceInstances #TechnologyManagementServices
- 1,032 Views
- Mark as Read
- Mark as New
- Bookmark
- Permalink
- Report Inappropriate Content
Hi @Sam Webb,
with CSDM 5 and the emphasis on different service instances and their relations we see a lot of opportunities to better model our services.
Could you please give your thoughts on why you have choosen the given relations between service instances and how those influence the impacted service analysis in incident management and change approvals in change management? These are two areas that make some headaches for us.
- Mark as Read
- Mark as New
- Bookmark
- Permalink
- Report Inappropriate Content
@Miklos Palfi fantastic question! In this instance, I haven't put too much thought into the Service Instance relationships, merely indicated that there are relationships.
I am planning an article in the next few weeks that will go into more detail on why we'd choose something versus the other, but in short:
- If it's hosted on (eg, Platform / Application) then think about "Parent Application Service"
- If it's dependent on, then think about "Depends on:: Used By" type relationships in the appropriate direction. Check the suggested relationships in Service Instance / Service Auto.
Hope that helps for now!
Sam
- Mark as Read
- Mark as New
- Bookmark
- Permalink
- Report Inappropriate Content
I will revisit and read this again properly, as I know its valuable! However for us over 50, can we make it a punishable crime to post any image in low res, such that you cant read it with young eyes, let alone knackered ones!
The flow image above, makes you waste precious seconds zooming in, in case 1)its just your eyes, 2) it may just be the render quality which may improve once clicked, 3) maybe you can read it if you squint with the left, or right or both, no? Write a post moaning about how full DDA Web 3 compliance or whatever the latest all inclusive accessibility regs are, are IGNORED when images are so poor.
Appreciate it may just be obvious to those that know, but its all learning and we are all at different stages.
In addition, thank you for bothering to share this in the first place, it is appreciate despite the rant and vent. And a big thanks to @Richard Hine for starting my day with joy and a moan (perfect combo!) 😮 )
Sharon
- Mark as Read
- Mark as New
- Bookmark
- Permalink
- Report Inappropriate Content
Interesting post @Sam Webb.
There is collateral for this already published by ServiceNow but I like to produce a terms of reference/data dictionary with each CSDM object, a precise plain-English description of what it is/does and how it functions within the framework, and a selection of real-world examples. It can take some stakeholders a long time to understand what each object does, and even more why it's necessary. Common/critical attributes per object also helps to prevent data being dropped into the wrong buckets.
This is also why I don't like the portfolio element being relegated to the Fly stage. With my customers I encourage them to think about portfolio design early on - by scoping out the business and technology management taxonomy layers it helps stakeholders to clearly understand what services they offer to customers (and in what categories), and what sort of technology management services they need to set up and define. Customers almost always have some sort of service portfolio data, even if it's just in a SharePoint site!
I've modelled TBM in ServiceNow Service Portfolio before (any other SM framework will also do) and with a few sample services I've found it makes understanding the upcoming stages of CSDM much simpler (as well as the critical need to consider your business app portfolio early on).
BR
Mat