I have recently started reading up a little on Cosmos DB and found that they have 5 consistency settings from strongly consistent to eventually consistent and everything in between.
Suppose I have chosen a strongly consistent setting. Now if I update the address for someone with a Social security number of 123-45-6789 and then I immediately try to read it, will it block me until it updates a majority of the copies? Or will it just give me old data until all the required number of copies have been updated?
As a follow up question, if it doesn't block me, what makes the latency go up for strongly consistent database?
Based on the clarification in the cosmos db consistency document:
Strong consistency offers a linearizability guarantee.The reads are guaranteed to return the most recent committed version of an item. A client never sees an uncommitted or partial write. Users are always guaranteed to read the latest committed write.
So,i think the latency is predictable following the CAP theorem.After all, strong consistency is at the expense of latency, ensuring a high degree of data consistency.
If your scenario is acceptable,Session Consistency Level is suggested by above link which is more balanced.It provides write latencies, availability, and read throughput comparable to that of eventual consistency but also provides the consistency guarantees that suit the needs of applications written to operate in the context of a user.