Fuck SQL

Fuck SQL

Attached: PQ3S41EWyYeT.png (2000x600, 29.46K)

Yeah, fuck actually being able to use querys in your db instead of just being able to fetch and store data with Id.

mongodb cpu and disk usage at idle and fresh install is insane

>t. filtered brainlet

I missed SQL so much when I had to use this.
I never missed mongo once I stopped using it.

Incorrect. Use mongo when no complex relational querying, use sql otherwise.

This

Why would you have data in 2 places? Probably code that uses different databases won't be interacting much.
Just use json columns in SQL if you need to store attributes that aren't relevant to queries (and be careful what you decide isn't relevant in a query because it may become relevant later)

can someone explain the difference between relational databases and shit like mongo/nosql? from what i understood, nosql is not concerned about data redundancy and normalization? the first time i asked this question, people just tell me "storage is cheap so who cares". is this the actual case for nosql? just make ID'd objects you can fetch and store without regard for storage, schema, or design?

non-free software

Basically you got it, but it nearly always ends up becoming relational data and either you migrate to SQL or pay out the ass for added CPU time and development cost working around the lack of relational queries. All the work SQL could do validating data types, relations and structure becomes application code, and requires very strict coding to make reliable.

>storage is cheap so who cares
No it doesn't work like that. If you had a UFC event, and you stored all of the fights in an array within an event object, you wouldn't be able to properly query things like "all of the fights that UFC fighter x had" as the fights are scattered all over the place in different objects and hard to get too or query.

To speak to this, json-schema is very popular in the nosql world to validate objects going in or coming out of the db.

>SQL
>NoSQL
>apples
>oranges
OP is a brainlet faggot.

Fuck SQL

Attached: fuckSQL.png (423x408, 71.09K)

so its not necessarily "storage is cheap" but "storage is secondary"? in comparison to relational databases using normalization to reduce data redundancy, nosql prioritizes proper fetching and storing of data objects in functions, etc. without throwing the entire concept of limited storage out the window just because of the presence of horizontal scale. do i have this correct? please be patient i am retarded

Storage is secondary yes. relational database doesn't just reduce redudancy, but makes things much easier to query. You can normalise mongodb, but then you just have to write all of the queries in application code, essentially reconstructing a sql dbms within your own code.

>Mongoloidb

ah i guess i can understand this use case. i am retarded so sometimes SQL queries are too hard to formulate for my small brain. but i see that nosql is a good solution to just having a good database system without the complexity. and then just using *easier to understand than sql* solutions to your application code. is this a common thought process when dealing with sql/nosql decisions?

People should use relational (SQL) databases in 95% of the cases.
And then in the remaining 5% think very hard and perhaps there is 1% of cases justifying using something else than a relational database.

There's tons of shit data that no one really cares about and that doesn't require any consistency (likes, comments etc.) NoSQL is perfect for that.