Geek Answers Handbook

Choosing a Database Technology

We are going to create an online platform (API, servers, data, Wahoo!). For context, imagine that we need to build something like twitter, but with comments (tweets) organized around a live event. Information about the live event itself should be delivered to customers as quickly and consistently as possible, while comments about the event can probably wait even longer. We will take care of the completion of the live event.

Scalability is very important. We want to start renting slices of VPS and scale from there. I am a big fan of the cloud and would like to stay there as long as possible. We will probably use ruby.

I am convinced that I want to try document storage instead of RDBMS. I like the idea of ​​no storage scheme and promises easier scalability, focusing on the key value.

The problem is that I do not know which technology is most suitable for our platform. I looked at Couch, Mongo, Tokyo Cabinet, Cassandra and RDBMS with blurry documents. Any help in choosing the right tool for this particular job?

+5
source share
3 answers

See a comparison of NO SQL BJ Clark alternatives .

Scalability is very important.

Then you need to consider excerpts from your blog:

  • -
  • Redis -
  • -
  • MongoDB - ( )
  • -
  • Amazon S3 -
  • Couch - (Clustering )
  • MySQL -

HyperTable. No-SQL. Google BigTable . , , Baidu Rediff.

:

, , , , . .

- Twitter. , Twitter Ruby , , Scala.

Ruby front-end. , , , Scala Erlang.

+7

. , Cassandra , Dynamo (, Voldemort Dynomite): , , /. Cassandra Twitter, Mahalo, Ooyala, SimpleGeo, WebEx (http://n2.nabble.com/Cassandra-users-survey-td4040068.html), Cassandra EC2 rackspace.

+1

( node), CAP.

http://www.julianbrowne.com/article/viewer/brewers-cap-theorem

It is not easy, but you need to choose, there is always some kind of compromise.

+1
source

More articles:


All Articles