Jow Forums approved databases?

Jow Forums approved databases?

Attached: ten.jpg (290x174, 5K)

Other urls found in this thread:

news.ycombinator.com/item?id=18442941
sqlite.org/codeofconduct.html
blade.tencent.com/magellan/index_en.html
twitter.com/AnonBabble

a csv file

>database.txt
>674,344 lines
>edit with Notepad

MS Access

Just keep it all in memory.

>not making the user remember everything in binary

LMDB for in memory key value
Postgres for everything else

>implying anyone on Jow Forums knows anything about databases

my diary desu

You mean like the one JEWS AND FEMINISTS KEEP ON STRAIGHT WHITE MALES!!!

PostgreSQL or SQLite3.

Attached: toriel with massive eyebrows.png (576x538, 62K)

influxdb

Hey, I know stuff about databases. I just don't know if I'd approve of any of them. You just kind of have to make your choice and commit to it. Or hire a 6-digit salary DBO to tell you which one will suit your needs best.

MongoDB is pretty great as long as you don't care too much about your data.

Oracle or SQL Server if we’re talking about a real DB. Postgres if you’re poor.

New version of MongoDB released just some months ago have all the ACID properties.

>tfw writing recursive mysql queries at work for my company's cms

i like programming

mnesia
actordb

Oracle

>Oracle
news.ycombinator.com/item?id=18442941
never.

Attached: Delayclose.jpg (380x247, 93K)

Oracle works just fine. Hire a good DBA that knows their shit and keeps things up to date.

>Oracle works just fine

Attached: 1435778390206.jpg (280x400, 56K)

t. someone who has never worked in a company with a giant Oracle DB

PostgreSQL

whats the source of your data?
how do you want to use it?
how many people need to have access to it?

Attached: 1441333469448.png (465x640, 501K)

MariaDB

Postgres is best gres.

go buy some almond milk hipster

what's the worst thing that can happen if someone knows what row your data is on? Is it a bad idea to use row number as an id?

this

Honorable mentions to sqlite3 and Berkeley DB
Most home gamers pick overkill solutions when sqlite3 and Berkeley are good enough. Berkeley DB is also pretty god damn powerful at this point, but people virtually never talk about it anymore because its attached to an "ancient old" name.

did you read that comment? the fact it works it all is a testament to code monkey suffering.
who the fuck in their right mind would pay money to larry ellison for that?

If you use anything else you are retarded

Attached: 1200px-Postgresql_elephant.svg.png (1200x1238, 120K)

>code monkey suffering.
Not my problem kid. I don't care about your code monkey's feelings. As long as the product continues to be as good as it is with maintaining databases with 100s TB+ data and great stability I could give not a single shit if Chin Chong Bin is upset about having to do his job.

t. someone who worked in a place with managers incompetent enough as to think buying Oracle was a good idea

Attached: GdaKOvB.png (645x773, 107K)

postgres if you're going big/complex
mysql is fine for everything else

Feel free to forward all complaints to Cerner Corporation. They are one of the top market share holders in eletronic healthrecord systems running your hospitals ;)

Oracle costs nearly $50k/cpu/yr. I doubt anyone here can afford that. Jow Forums approved databases should be free to use ones only. No corporate/enterprise swank.

yeah dude we all know how great the technology decisions made by industries that are under regulatory capture are
you sound like some IT monkey retard that doesn't even know how any of this shit actually works and can only spout third-hand marketing drivel

Attached: 1464504982266.jpg (1023x682, 139K)

You probably also hate IBM and POWER i bet. I hope someday you'll get a good job at a good company that knows how to manage databases and experience how wonderful Oracle is when properly maintained.

log based architectures

nice trips trips nice trips nice BASED

shut the fuck up retard

This

These.
Postgres is the ONLY universally accepted Jow Forums datatbase.
...and yes, it's fucking badass.

And yes, I do have a PostgreSQL tattoo

Tin can and wire to the chinks you keep in your basement

gdbm

If you know what you're doing, have buckets of money and you're competent, MS SQL or Oracle.
If you know what you're doing: Postgres
Anything else is just absolute retard.

SQLite for toy programs (or just anything that's single-user).
PostgreSQL for anything else.
Also, not Jow Forums approved, but MSSQL is nice if you ignore the PROPRIETARY, and the fact you need a lawyer to understand how much you're supposed to pay for it (pay per concurrent connected user, per core, per server, per available feature-set, etc.).

PostgreSQL for big, sqlite for small things

Why is Postgre so good?

Kys

And no, you cannot see it. It's for the ladies' eyes only- and even then they have to demonstrate that their schema complies with 5th normal form beforehand. Nothin personnel kid.

What if your program saves some Ids for whatever reason, and then you delete an entry. All your id's are now fucked up

first for elasticsearch !

Attached: 642.jpg (605x727, 57K)

Attached: DATA BASE.gif (480x270, 752K)

postgres, oracle or MSSQL for real world enterprise
mariadb for medium projects
sqlite for things that just need a small embedded db
redis for caching
stop projecting and gtfo

apache cassandra

MySQL, Postgres, MariaDB are all fine, Redis for caching.

Based DB2

Riak

Cassandra

u trollin'?

I hate hitchhiker thumbs.

sqlite

Attached: rs8fkinmhrt11.png (986x2143, 294K)

>database.exe

SQLite got rid of that code of conduct and replaced it with the crappy Mozilla one.

>tfw migrating my company's dbs to vitess because mongo is an unscalable piece of shit

It's still linked right at the top of the page though which I think is funny enough
sqlite.org/codeofconduct.html

json

Lol

>God Tier
Comma Separated Values (CSV). It doesn't even have to be a comma, it can be white space, or any other character you want. There's zero bloat, high readability and portability, almost all devices natively support this format, and its extremely easy to implement.
>OK Tier
Binary files. Why waste space? Put the bytes down directly. Variable header, metadata, block size, signature. Very easy to implement, allows portability through encoding to base64 for services which do not allow all data values. Can be read with hex editors. Beware of endianness.
>Pajeet Streetshitter tier
Anything else.

Attached: c32.jpg (640x638, 31K)

S3 never

at least the LNs are finally continuing after a 2 year hiatus

>Comma Separated Values (CSV)
CSV isn't 100% portable and isn't the simplest of formats since storing special characters in strings has its limitations. How would you store commas in strings? How would you store double quote characters? How would you store line endings within strings? There are fairly standard ways of doing all these things but they're not universally supported and make CSV a bit more complicated than you'd like to think. Also there's the issue of different character encodings and the dreaded UTF-8 BOM.

Your suggestion of "binary files" doesn't give any kind of standard format for storing data at all (unlike messagepack / protobuf) but in any case, a database is more than just a format that data is stored in. Databases typically provide ways of looking up data quickly with indexes, as well as manage concurrent updates to data. If you're using files to store your data, then the database is really your filesystem and not the files themselves. Filesystems can be a lot more useful for organizing data if you make use of directories for faster access to your data rather than storing all your data in large binary / CSV files. If you have one large CSV file with millions of rows and lots of columns, getting a list of rows where a particular string matches is a very slow process.

So much csv in here. Only one mysql :(

Csv is plebian. For big projects you’ll just end up rewritting a shitty version of a bonafide db.

+1 cookie

I do all my data like this when programming. Should I consider learning an actual database? One of my projects recently produced over 30gb of csv

Yes you should and you are retarded. You can't ensure transaction safety unless you lock the entire file on a write operation,making your data layer either extremely slow or very inconsistent. There's absolutely no excuse to not at least use SQLite and using csv files as a fucking database is "I'm a hacker because I code in notepad"-tier retarded. Learn a database or kys fucking faggot. You're not cool and not minimalistic by "using CSV as a database", you're just homosexual and probably also unable to learn new things if you really took the meme that far to 30gb

>blade.tencent.com/magellan/index_en.html
Turns out power of christ can't replace good test coverage

>Why is Postgre so good?
Fast, enterprise scale, open-source, cross-platform.
Can do most things Oracle can do, but free.

Postgres can scale but it isn't enterprise scale. It's a great database for 99% of RDB work.

>Only one mysql :(
But that's good, MySQL is utter shit and you have no excuse for using it when Postgres exists.

>UTF-8 BOM
>byte order mark for encoding that fits into a single byte
another technological marvel brought to you by microsoft "engineers"

>thinking datastructures should have methods
oh my

Are you referring to protobuf?

Came here to post this.

Fucking tell me about it.

The support engineers at my work wanted me to add a UTF-8 BOM to a csv export process.

I laughed and told them no.

clickhouse

Impala Hadoop clusters

>30gb of csv

pajeet wash your hands.

Where the non-relational DB’s at?

SPLUNK

obviously yes, mysql is simple anyway

MySQL vs Postgres?

As an Oracle DBA, fuck Oracle. My job is pretty much reverse engineering shitty errors and bad programming for a 10GB mess of binaries. But the salary is sweet at least.

Oracle has a free-ish version, it's a piece of shitty but there is one.

postgres is great.

sqlite is also pretty nice, if you can get away with it. Most of the time, you actually can.

What's wrong with mySQL?

Based and redpilled