Reputation: 18961
Hi as I am new to Cassandra I am reading some books, docs, and blog posts to get myself familiar with it. Now I would like to get some answer to how to implement "Accumulating Materialized Views In Cassandra" and by that I mean : Suppose that we have the following model Game { userId, gameName, time,gameMoney .... , etc. } If I want to "Aggregate" games by userId, gameName, I will need denormalize the table to some Materialized View, where the same data is stored but in different model.
What I want is when some record is inserted into the first table a record is inserted in the materialized view or it is update and accumulated if exist.
For example :
Game { 1,gameOne,12:00,1.2}
Game {1, gameTwo, 12:00, 2.3}
Game {1,gameTwo,11:00, 2.3}
In the denormalized Materialized View, I want to have two record ( an aggregation by userId and game name) that will look like
UserGames { 1,gameOne,1.2 }
UserGames {1,gameTow, 4.6 }
I know about triggers but curious if there are some other ways to accomplish that ?
Upvotes: 3
Views: 565
Reputation: 852
Aggregating materialized views are still not implemented. Here is the relevant Jira ticket and a video explaining how the feature would look like.
Depending on how much data you'll have, you could consider using aggregates.
Upvotes: 5