Hi team,
I have a multitenancy use case where dat...
# troubleshootingm
Miguel Ángel Fernández Fernández
04/12/2023, 11:47 AMHi team,
I have a multitenancy use case where data is ingested by kafka and stored in cassandra. Depending on the data, these are stored in different keyspaces (Multi-tenancy with shared database and separate schema).
We currently have an implementation that incorporates multiple cassandra sinks (one for each keyspace), but this implies multiple Cassandra connections.
Looking at the documentation, we have observed that the sink can be configured with a DefaultKeyspace. Is there the possibility of being able to modify that keyspace for each tuple? In this way, we would not need multiple connections with cassandra.
Miguel Ángel Fernández Fernández
04/12/2023, 1:23 PMWe are studying other sinks and there is always a similar pattern: static queries, which points to a table (JDBC) or a collection (Mongo).
This suggests that perhaps a better approach would be to deploy a topology for each tenant. Unfortunately, we would have the same number of connections and also OPS would be more complicated.