Question-: when you use the clustering columns A. While writing the data, memtable also sort the data based on clustering column. B. While writing the data, commit log also sort the data based on clustering column. C. Based on the clustering column new commit log files are created. D. While reading the data, engine would always get the data sorted by clustering column.
Answer: A, D Exp: Clustering column is mainly used for pre decided sorting order. When you write the data, it is always sorted based on the clustering column. When written to Memtable. However, please note that Commit log does not have to do anything with the Clustering or partition key column. It always adds the data in sequence. While reading the data, it would be always sorted.
Admin Only
Question-: Select correct statement with regards to Chunk cache and OS cache. A. OS page cache is sized dynamically by the OS. B. Chunk cache is sized dynamically by the OS. C. OS page cache size should be predefined and remain static. D. Chunk cache size should be predefined and remain static. E. You should always keep the chunk cache very high.
Answer: A,D
Explanation: Chunk cache is used to cache the small parts (chunks) of SSTable files so that read operation faster.
OS page cache is sized dynamically by the operating system, it can grow and shrink depending on the available memory. And chunk cache should be sized statically. Suppose you have decided to keep the chunk cache too small then available server memory will remain unused.
Keeping the chunk cache very high can also cause that your Cassandra server process can be killed by the OS. Whenever you change the chunk cache size, you need to restart the Cassandra process.
Admin only
Question-: Which of the following is true for the Cassandra clocks? A. Cassandra uses the Network Time Protocol to synchronize the clocks on all nodes and application servers. B. Cassandra always uses the local time stamp of the node for storing timestamp with the data. C. Cassandra always uses the UTC time stamp of the node for storing timestamp with the data. D. Cassandra recommend that you install the either Google or Redhat clock service on the Node.
Answer: A,C
Explanation: Cassandra uses the NTP Network Time protocol to synchronize the clocks on all nodes and application servers. And to create the time stamp it always uses the UTC. If you want you can convert this value as per required time zone.