Why CouchDB doesn't scale

http://spyced.blogspot.com/2008/12/couchdb-not-drinking-kool-aid.html

by jorgeortiz85 • 17 years ago

50 30 17 years ago

spyced.blogspot.com

bitdiddle4• 17 years ago

Wasn't this posted and discussed some time ago? It is somewhat dated, and the arguments weren't very convincing, in fact they struck me as typical marketing FUD.

Sure the Postgres folks have years of experience and so forth but it's not like the literature is not available to be incorporated in new approaches. CouchDB hits a certain sweet spot in combining a few new ideas together and the folks hacking it are some of the best. I highly recommend giving it a hard look.

I also hear that the canonical folks will be bundling it in an upcoming Ubuntu release so it will be easier than ever to check it out

Confusion• 17 years ago

If you see how long it took for MySQL to get features like views and stored procedures, which I consider pretty much indispensable, then warning people that something like CouchDB is far from being mature is hardly FUD. The worst thing is: the cool crowd doesn't even know what they are missing. It's like talking to someone using MySQL 7 years ago: they'd be all enthusiastic, not understanding why you would worry about the absense of mundane things like views or, god forbid, the ability to use multiple indices per table in a query...

aurumaeus• 17 years ago

I've used every major relational database on the market, and views & stored-procedures being requirements are blanket generalizations that are arguable at best. Both of them are enterprise tools for working around stupid schemas and bad code.

Arguing that DBAs 'require' the complex features in the other relational database is like saying that Macs 'require' more complex interfaces so IT people can work with them.

Backwards, and wrong.

trezor2• 17 years ago

Yeah. It's pretty much like MySQL all over, except this time they are doing something new which doesn't have any absolute, proven references out there, so you have to argue on a philosophical level why they are on the wrong track instead of citing absolute metrics.

The sad thing is that these people use MySQL of all things as a reference point as what a relational database can do, and hence to them it seems all reasonable.

evgen• 17 years ago

> The worst thing is: the cool crowd doesn't even know what they are missing.

Actually, they do. A lot of the people interested in and writing code for these new databases are the ones who have seen first-hand the failures of the traditional DBA view of the world. If you look at where a lot of these projects are coming from and who are sponsoring them you will see that they are some of the companies that are at the leading edge of internet scaling and deployment. The advantage they have which you seem to ignore is that they can look back at the wrong turns in the development of the current crop of RDBMS and avoid those particular dead-ends.

dkarl• 17 years ago

MySQL was a relational database that was missing some very important RDBMS features. CouchDB is not a relational database, and the developers know it's not.

What it is Not

A relational database. A replacement for relational databases.

(http://couchdb.apache.org/docs/intro.html)

Obviously, CouchDB is a cool thing for the cool crowd, and doubtlessly many of the cool crowd are going to shoot themselves in the foot by using CouchDB when they should be using a lame boring RDBMS, which they can't bring themselves to do because that would be like wearing a polo shirt with their company's logo on it. That doesn't mean CouchDB is a bad idea. It's just a very sexy idea that will break a lot of hearts.

gaius• 17 years ago

If you see how long it took for MySQL to get features like views and stored procedures

I was using MySQL back in the 90s, and I remember their documentation. You didn't need transactions, they were slow - do it in your own code if you had to. You didn't need foreign keys, they were slow, enforce integrity in your code if you had to. You didn't need stored procs... You get the picture.

Baked deep into MySQL's culture is "if we don't have it, you don't need it" - even now. Feature-wise MySQL is probably 15 years behind Oracle - yet somehow its legions of fans are convinced it's on the cutting edge!

igorgue• 17 years ago

> I also hear that the canonical folks will be bundling it in an upcoming Ubuntu release so it will be easier than ever to check it out

I think you're talking about "Ubuntu One"[1] a file syncing tool (kinda like Dropbox), and Ubuntu One use CouchDB for their storage.

[1] https://ubuntuone.com/

edrtghuj3• 17 years ago

>The reason PostgreSQL et al have those features is because people want them.

The reason those features are in PostgresSQL is because they are in SQLserver, ... because they are in Oracle, ... because they are in DB2.

Based on that logic C should have had a report generator and the unix filesystem should have records - and your car would have a saddle and whip holder.

aaronblohowiak• 17 years ago

Only those cheap imports don't have whip holders.

Otherwise, your contention in-domain feature copying somehow translates to cross-domain feature transfer is something that I fail to understand. Can you explain it better?

edrtghuj3• 17 years ago

Postgress has certain features because it is competing, both technically and in marketing check-boxes, with other SQL databases. CouchDB is not an SQL DB (and not even an RDMS) and so doesn't automatically have to have the same rich set of features.

james2vegas2• 17 years ago

No, it has those features because it aims to be SQL standard compliant, as do those other RDBMSes.

Where is the Standard for these 'Document-oriented' DBs?

jrockway• 17 years ago

CouchDB doesn't scale, but not for any of the reasons the author of this article lists. I especially like meaningless sentences like, "just like PHP programmers don't get namespaces." While true, it's not "just like" that.

kscaldef• 17 years ago

You keep using that word. I do not think it means what you think it means.

chibea• 17 years ago

Not talking about immutability at all when talking about CouchDB's (potential?) scaling ability is odd.

Even if immutability in DBs was no innovation from CouchDB itself, CouchDB will have raised the level of awareness for immutable DBs and probably drive further development in this area. IMO immutability is the very basis of creating scalable applications. So, when talking about CouchDB and scalability you should talk about why the current CouchDB implementation is not scalable in spite of immutability.

Disclaimer: I've only read about CouchDB and never actually used it. I'm not convinced of every feature of CouchDB, I don't think, for example, that it is a good approach to dump type-safety in dbs altogether, though, SQL and RDBMS are not the last word in databases and it is valueable to test other, new models of data storage.

Why CouchDB doesn't scale

Related Stories