Cassandra uses a ring architecture where:

• All nodes are arranged in a logical ring
• Each node is responsible for a range of data (token range)
• Data is distributed using consistent hashing
• No single point of failure as there's no master node
• Nodes communicate using gossip protocol
• Data is replicated to multiple nodes for fault tolerance

Each node knows about every other node in the cluster and can handle any request.

Tombstones are markers that indicate deleted data. Since Cassandra is distributed:

• Deletes can't remove data immediately from all replicas
• Tombstones ensure deleted data doesn't reappear during read repair
• They have a configurable TTL (gc_grace_seconds, default 10 days)
• Too many tombstones can impact read performance
• Compaction eventually removes tombstones and associated data

Replication provides fault tolerance:

• Replication Factor (RF): Number of copies of each piece of data
• Replication Strategy: Determines which nodes get replicas
  • SimpleStrategy: For single datacenter
  • NetworkTopologyStrategy: For multiple datacenters
• Coordinator Node: Any node can coordinate a request
• Replica Nodes: Nodes that store copies of the data

Example: RF=3 means each piece of data is stored on 3 different nodes.

• Partition Key: Determines which node stores the data (distribution)
• Clustering Key: Determines the sort order within a partition (organization)

CREATE TABLE events (
    user_id UUID,
    event_time TIMESTAMP,
    event_type TEXT,
    data TEXT,
    PRIMARY KEY (user_id, event_time, event_type)
);

• user_id is the partition key
• event_time and event_type are clustering keys
• All events for a user are stored together, sorted by time and type

Cassandra supports various data types:

Primitive Types:

• TEXT, VARCHAR: Strings
• INT, BIGINT, SMALLINT: Integers
• FLOAT, DOUBLE, DECIMAL: Floating point
• BOOLEAN: True/false
• UUID, TIMEUUID: Unique identifiers
• TIMESTAMP: Date and time
• BLOB: Binary data

Collection Types:

• SET<type>: Unordered unique values
• LIST<type>: Ordered values (duplicates allowed)
• MAP<key_type, value_type>: Key-value pairs

Advanced Types:

• User-defined types (UDT)
• Tuples
• Counters