IG1157 Digital Platform Reference Architecture Concepts and Principles v8.0.0
For now I can share a couple of cases I challenged before.Last weeks, for the glory of the reusability, we started to share already transformed service with other product team which they physically positioned out of our OCP ecosystem.So, they started to consume our resources, RAM, CPU etc. From the backend perspective, they joined to one of our big flow in the middle. At the end of the flow, there is one data transformation remote service and that remote service owners need to know which channel pull or push the data and I want to know who is the consumer of our resources service by service. (I wonder If we can see that on the OCP dashboard.) Besides, I want to parse logs in our logging system according to some kind of channel ID.The other topic is trace ID. Recently, I had clarified that the details of wishes "don't share data in the header" is actually both archive sizing and security issue. From the security perspective, we are at the safe zone if we use it as a header property, however data teams challenged with the government regulative rule of "you must archive all request data for six months". The data teams want to minimize request logs which included header. So, they don't want their logs to be crowded.
If we can use polymorphism in request models without broke any TMF rules I would like to use it.
And maybe there are other cases I don't know yet, I will ask my team.