Discuss the role of data replication, partitioning, and sharding in achieving high availability and fault tolerance in distributed databases, and analyze the trade-offs between consistency models, such as strong consistency, eventual consistency, and causal consistency, in distributed data management systems.