Cassandra Practice Test

How many rows are recommended as the upper limit for a partition in Cassandra?

1. 5000 rows

2. 100000 rows

3. 10000 rows

4. 1000 rows

The correct answer is: 100000 rows

In Cassandra, the recommended upper limit for the number of rows in a single partition is generally considered to be around 100,000 rows. This recommendation is based on the need to maintain efficient read and write performance while avoiding potential issues with large partitions. When partitions grow too large, they can lead to performance degradation, increased latency, and challenges in data management. Large partitions can also complicate compactions, increase the likelihood of out-of-memory errors, and make data distribution across nodes uneven. By keeping partitions within this recommended limit, users can ensure better scalability and maintain the performance characteristics that Cassandra is designed to provide. Other options suggest fewer rows, which may offer better performance for specific applications with lighter load requirements, but the guidance of around 100,000 rows is a balance between performance and scalability for many typical use cases in production environments. This allows for a manageable load during operations like reads, writes, and repairs, leading to a more stable and efficient database experience.