[leif]

If you were using upserts I expect you were having problems due to the optimizer retrying all possible plans (including table scan) periodically. This is reflected in https://github.com/Tokutek/mongo/issues/796 and is fixed in 1.4.0. If you'd like to try another evaluation, get in touch with us and we can help you track down whatever problems you see.

Not all mongodb code will optimally use tokumx without any changes. Concurrency is hard and mongodb encourages some patterns that are bad for any concurrent database. For example, count() for an entire collection is not, and could never be, as cheap in a concurrent database like tokumx as it is in mongodb.

[rogerbinns]

Thanks for the offer, but the mongorestore times (against MongoDB) being over a week makes this too risky.

The code making changes was insert (mostly) with a few upserts, but the latter was by _id. My hypothesis as to the cause is that tokumx adds implicit transactions and then there are some arbitrary restrictions around those transactions (eg how many outstanding at once, timeouts in lock acquisition) and after a few hours one of those was hit. The error message was something about being unable to start a transaction.

> Not all mongodb code will optimally use tokumx without any changes

The goal wasn't to be optimal or anything like that. It was initially about space consumption (where you did really well) and verifying the same client code ran correctly. We have two setups so one would run toku and one mongodb and data processing results compared.

[leif]

Ok. Well, you said you were waiting for it to be ready, and I think it is. We'll be here when you get a week free to tinker.