3 ms·
It sort of looks like Cassandra is where all the people with A LOT of data are going. Any particular reason(s) for that? (Edit: as compared to other 'non trad
by davidw 17y ago
It sort of looks like Cassandra is where all the people with A LOT of data are going. Any particular reason(s) for that?
(Edit: as compared to other 'non traditional' database systems).
- mrduncan 17y agohttp://en.wikipedia.org/wiki/Cassandra_(database) http://en.wikipedia.org/wiki/Cassandra_(database) has a pretty good overview if you're not familiar with Cassandra. Cassandra only guarantees eventual consistency where as most RDBMS systems are focused on always showing a consistent view of data. Because of this (and other trade-offs), Cassandra is able to provide much higher IO rates as well as improved fault tolerance and linear scalability. Twitter, Digg, and Facebook all happen to have usages which fit perfectly with this model. If you see a status update a few seconds (or even minutes) late occasionally it's no big deal. If your bank funds aren't matching for that long, it can be a big deal. Edit: I just started looking into Cassandra a few days ago so someone please correct me if I've mis-stated anything.
- davidw 17y agoI'm more interested in the comparison with Redis, MongoDB, et alia.
- jbellis 17y agoRyan King gave a good summary in his interview. (http://nosql.mypopescu.com/post/407159447/cassandra-twitter-an-interview-with-ryan-king http://nosql.mypopescu.com/post/407159447/cassandra-twitter-...) Basically, Cassandra has the best answers to the important questions: How will we add new machines? Are their any single points of failure? Do the writes scale as well? How much administration will the system require? If its open source, is there a healthy community? (And I would add, is it _fast_ as well as scalable?)