Does it make sense to continue looking for comparison table where I could see the difference between completely different DB types?

Question

I would like a comparison table where I could see the difference between completely different DB types.

I may be incorrect to desire it because all DB types have different nature but I'm interested in only aspects like latency and RPS for CRUD operations for: PostgreSQL (RDBMS), Clickhouse (time-series), Radis (key-value), MongoDB (document), Neo4j (graph).

Does it make sense to continue looking?

Answer 1

I am not sure that it is correct to compare document database (MongoDb), relational database and other types of databases.

Because these types of databases have various goals and under the hood have different data structures. What I mean is this:

• relational databases use B+ trees,
• In MongoDB, data is stored in flexible, JSON-like documents called BSON. It allows for dynamic and nested data structures, making it more suitable for handling unstructured or semi-structured data.
• Redis primarily uses key-value data structures to store and search data. It also supports other data structures such as lists, sets, sorted sets, and hashes.

So it seems that your question can be asked in another way like this what the differences are between data structures.

In addition, you can read this great article about MongoDB vs PostgreSQL: A Comparison in Brief.

UPDATE:

Anyway i would like to see some numbers. Lets say i know that Clickhouse is not good for updating single entries because of its nature. But it does allow to do it though very slowly. So i want to know how slower it does that comparing to PostgreSQL for example.

These numbers varies because foreign keys, key constraints, indexes, query plans, table locks and also hardware play crucial role in performance. However, ClickHouse does not lock tables. Read more about ClickHouse vs PostgreSQL

Answer 2

Actually it more or less seems like i was looking for Performance database ranking for SQL, NoSQL, and NewSQL