Why NoSQL is better for big data applications

projectpro.io

7 pointsnigamanth7 comments

This article makes the classic mistake of conflating SQL with RDBMS. As such it is unlikely to make any interesting points. Sniff test failed.

What's lost in doing that though? I'm not a DB expert, but from my experience SQL and relational databases pretty much go hand-in-hand.

SQL can be used as a query layer outside of an RDBMS. Spark is a good example of this.

RDBMS as a class of databases has nearly as many variations as “NoSQL”.

So presenting SQL and RDBMS as the same thing or implying NoSQL is a single thing is oversimplifying to a point that any interesting nuance is lost.

You need to pick a database that suits your need but to make that choice you’re going to have to look deeper than “RDBMS vs NoSQL”.

Look at concepts like key-value stores, document databases, graph databases, shared nothing architecture, shared everything architecture, data normalization, CAP, and ACID. This is where the meaningful tradeoffs happen.

Site isn’t loading for me, anyone else having issues?

Edit - looks ok now! Maybe just my connection.

It is loading fine for me.

Edit: Oops autocomplete typo!

they should have used SQL.

You're not missing much, its packed full of buzzword ads.

Reply on news.ycombinator.com