Maybe we can take a step back and ignore the ArangoDB specifics for the moment. I'm also organising NoSQL conferences and consulting NoSQL in general.

In T88549#1018451, @Fceller wrote:

There are thousands of indexable attributes, but not each document has every one of them - typical one would have tens, rarely up to a hundred. We need to be able to search by any of the attributes though.

I see. That could be the root cause of the problem. The test dataset that you used, did it contain a lot of documents with unset attributes for the hash-index? Currently we do not support sparse indexes - but my colleague has started enhanced to allow them. I'll check with him.

There are thousands of indexable attributes, but not each document has every one of them - typical one would have tens, rarely up to a hundred. We need to be able to search by any of the attributes though.

@Smalyshev May I ask how much main memory your server has?

Hi, I'm the CTO of ArangoDB, so my comments are most certainly biased. I still would like to tell you about our opinions on the raised issues, namely full-text indexes and blueprint.