Data Holders have expressed concern that current polling requirements to check ADR statuses every 5 minutes may not scale into the future.
Initial discussions with ACCC, on potential direction for the Registers ADR endpoints in relation to ensuring DHs update their internal caches on a timely basis, resulted in:
Our goal is a distributed cache with non-transactional APIs wherever possible (which reduces operational reliance on the Register)
This means using the meta data refresh API for suspensions, etc, which also means building a trigger mechanism
We would then deprecate (over time) the status APIs
We would also remove the five minute call requirement
We would add a tag based delta mechanism to ADR endpoint
We would then use this delta pattern for all endpoints (eventually)
A business process to trigger meta data refresh API would need to be developed
Data Holders have expressed concern that current polling requirements to check ADR statuses every 5 minutes may not scale into the future.
Initial discussions with ACCC, on potential direction for the Registers ADR endpoints in relation to ensuring DHs update their internal caches on a timely basis, resulted in: