Hi.
Any on-premise application (let’s call one here ABC) is typically hosted on one or several physical or virtual servers (let’s have here 2 virtual servers SRV123 and SRV456). These servers have often but not always the same OS (let’s say here Windows Server 2019 for both SRV123 and SRV456) and different purposes with different components (let’s say here SRV123 is the web application server with IIS 10.0 and SRV456 the database server with SQL Server 2019). They also run on different pieces of hardware or a single one (let’s say here that SRV123 and SRV456 run on 2 different Dell PowerEdge R550 Rack Server units).
In my example, Application ABC would have relations with the following IT Components in LeanIX:
- Software: Microsoft Windows Server 2019 + Microsoft IIS 10.0 + Microsoft SQL Server 2019,
- Hardware: Dell PowerEdge R550 Rack Server.
But SRV123 and SRV456 servers actually appear nowhere in this model:
- Of course, they’re not Applications,
- They could be IT Components, but none of the standard subtypes apply and there’s no relations with other IT Components to model their hardware/software composition.
In V3 metamodel, there was the optional Tech Platform fact sheet type, that had exactly this purpose of modeling compositions of IT Components to build Applications upon. In V4, the Platform fact sheet type has a more strategic purpose driven by Objectives, Initiatives and Business Capabilities, and thus does no longer fulfill this more granular technical composition purpose.
So, because I guess it’s a very common use case, what would be the best V4 practice in modeling such virtual (or physical) servers that host Applications:
- Extend the IT Component fact sheet by adding missing subtypes and relations with other IT Components?
- Still enable and use the optional V3 Tech Platform fact sheet in addition to V4 Application, IT Component and Platform fact sheets?
- Create my own new fact sheet type, but then why is it no longer part of V4 metamodel?
- Do nothing because LeanIX is not a CMDB, but then why should we be denied the opportunity to map in LeanIX diagrams these important core architecture nodes?
Many thanks.
