Is it true what ravendb says?

I have read these article and they say this:

RavenDB outperforms Couchbase by orders of magnitude when using queries under load** . Couchbase proved fragile in production, with high overhead on failure conditions. RavenDB’s failure model, on the other hand, proved resilient and allowed much higher operational flexibility and peace of mind.

https://ravendb.net/wp-content/uploads/2021/04/Couchbase-vs-RavenDB-Performance-at-Rakuten-Kobo.pdf

I use couchbasedb and I am exciting about it but is it true what ravendb says ? That is not a marketing advise or something else, I love couchbase

The report was written by ravendb so it might be a little one-sided. The comparision is interesting, but unfortunately they used the wrong couchbase api to retrieve documents by id, and they generate a primary key for couchbase which is not used (and also not recommended) which wastes a lot of space. They also use the free Community Edition instead of Enterprise Edition Enterprise Edition vs Community Edition vs Couchbase Capella . With a few minor adjustments, they could have produced a much better comparision. It’s unclear why they chose not to.

1 Like

On top of what Michael said, this was written 3 years ago. Since then a new storage engine has been released (Magma), which is a lot more stable and very performant for case used in the comparison.

Honestly, the comparison was built to make Couchbase fail. They decided to do everything you should not do with Couchbase, but pick the a suited strategy for RavenDB.

2 Likes

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.