Hi Arko
Firstly be aware that the API contract says nothing about your internal implementation of a resource inventory. The API is a logical exposure, and the implementation might be done by cards in a filing cabinet (although the performance might be rather challenging π)
However, you are right that retrieving a resource by ID when the persistence is spread over multiple stores might be difficult. Depending on your performance and cost requirements, you could have a redirection table that has ID as a key and a pointer to the underlying storage as a value.
But if the consumer knows the resource type, maybe it also knows unique values that are not ID, e.g. value of logical resource or serial number of physical resource, and use GET /resource without the ID.
------------------------------
Jonathan Goldberg
Amdocs Management Limited
Any opinions and statements made by me on this forum are purely personal, and do not necessarily reflect the position of the TM Forum or my employer.
------------------------------