Sorry, but a generic data system which allows all players to update the same data in a consistent manner so that it's guaranteed that any player's update will immediately be visible to all other players would be incredibly expensive to operate, and is not part of our current offering.
At a high level, there are two primary types of data in an online system:
1. Data which can be read by all players
For this, we have systems like Title Data, Publisher Data, Content, and Title News. These systems are distributed - sharded and cached - so that each element can be read by many players at the same time.
2. Data which can be written to consistently (write something, and it can immediately be read back) by players
For this, we have all the forms of user data, shared group data, etc. It's data that can be written to by the user, but because each entry is a single row in a data table, you cannot have a lot of players trying to read from it at once (it's just the reality of physics - each request must be serviced separately) and if multiple players tried writing to it at the same time, you'd have collision issues.
Now, there are specialized data sets that do allow for input by many players at once - our leaderboards (statistics) system is an example of this. For a data system which requires input from many sources and which can be read by many clients, a custom implementation is required so that it can meet the specific requirements of the design while minimizing costs. For example, in most (if not all) realtime online games where players can see each other moving, clients get updates for other player positions based upon a "relevance" calculation, which usually combines things like distance from the local user, whether they're within the view frustum, whether they recently interacted with the local user, etc.
1 Person is following this question.