Consistency and availability

Consistency and Availability

1. Consistency

Meaning

Consistency means that all users see the same data at the same time after an update.

In distributed NoSQL databases, full consistency at all times is difficult, so systems use tunable consistency.

Consistency in Cassandra

Cassandra provides tunable consistency, meaning the user can choose the level of consistency required.

Common Consistency Levels

ONE – Data written/read from one node
QUORUM – Majority of replica nodes respond
ALL – All replica nodes respond

Advantages

Flexibility to balance performance and accuracy
Suitable for different application needs

2. Availability

Meaning

Availability means the database remains operational and accessible even if some nodes fail.

Availability in Cassandra

Uses peer-to-peer architecture
No master node → no single point of failure
Data is replicated across nodes
Requests can be served by any available node

Benefits

Continuous uptime
Automatic failover
High fault tolerance

3. Consistency vs Availability (CAP Theorem)

CAP theorem states that a distributed system can provide only two out of three:

C – Consistency
A – Availability
P – Partition Tolerance

Cassandra follows

Availability + Partition Tolerance (AP)
Consistency can be adjusted as required

4. Comparison Table

Aspect	Cassandra
Default consistency	Eventual
Consistency type	Tunable
Availability	Very high
Failure handling	Automatic

5. Real-World Example

Social media likes count
Sensor data collection
Messaging systems

These applications prefer availability over strict consistency.

Comments