Why RDBMS (SQL):
- Frequent CRUD transactions for a limited similar type of data in GBs.
- It’s good for structured data.
- This kind of database is tightly structured with schema and perform slower (low latency) with huge growing data.
- RDBMS performs faster for low amount of data ( in GBs).
- SQL DBs- Oracle, MySQL, SQLServer etc.
Why MongoDB (No-SQL):
- Write once and read many for unstructured data.
- It’s faster than RDBMS for growing data on clusters/cloud in TB, PB etc.
- If there is a requirement to not update DB frequently (not mission critical), dissimilar data , then go for this.
- No SQL DBs- MongoDB, Cassandra, NeoJ, CouchDB, Hadoop, Cloudera, MapR etc.
In nutshell, if you have simple data without any complex relationship then why to choose complex RDBMS. Both are not replacement of each other, both have their own importance. I think it will be helpful to understand use cases of these two databases.