Question-: If you are using Hard Disk for the Cassandra node to keep it less costly, then what alternate thing you should do, so that Cassandra cluster perform better? A. Keep lesser RAM B. Get as much RAM as possible C. Disable the GC D. Compaction process should run only once in a day.
Answer: B
Explanation: At the first place you should avoid the using Hard Disk and prefer the SSD. But if you still want to keep the HDD, then keep the RAM as much as possible. So that more data can be cached.
Question-: If you have more RAM then it could A. Have better read performance by caching the data. B. Memtable would have to store less amount of data C. Memtable can hold more amount of data in memory D. Memtable would be flushed less frequently
Answer: A, C,D Exp: Having more RAM means you have more memory available for both the Cassandra as well as to Kernel. If you have more memory you can cache more data. Having more caching would certainly improve the read performance.
Similarly, Memtable is an in-memory data structure, which holds the recently written data in cache. As much as memory available it can have more data in memtable. And frequency of flushing the memtable would be reduced.
Question-: Which of the following hardware settings are good for Cassandra production cluster? A. Use 16GB RAM for each node B. Use the 4 GB RAM for each node C. You should use the 16 Core CPU D. Keep your node OS bind with the separate NIC (Network Interface Card) E. Bandwidth should be greater than 1GB/sec
Answer: A, C, D, E Exp: Yes, you should have as good hardware as possible for the Casandra cluster. Recommended ones are as below. - Keep the RAM between 16GB to 64GB on each node. - You should use the 16 Core CPU machine. - Bandwidth should 1GB/sec or higher - You should bind the separate NIC to your Cassandra cluster.