Loïc Lecrenier

An update on the progress so far: First, there is the issue of rate-limiting + search request timeout. This is probably the most important issue if we want to make...

Closing this issue in favour of https://github.com/meilisearch/meilisearch/issues/3547 , which supersedes it.

Fortunately this is just an issue within the tests. It comes from a simulated failure that is wrongly named `CorruptedTaskQueue`. We'll rename it to something clearer in the future.

A data point on the performance of the iterative search algorithms when dealing with mid-size documents (containing a few thousand words): Searching for `I love` in a dataset I created...

Another data point, this time from a user: > Hello! I'm doing some experiments with Meilisearch v0.30.4 and am surprised with how long queries are taking (~500ms to search through...

Fixed by https://github.com/meilisearch/meilisearch/pull/3542

Hi @vaishnavi-ramanujapuram, that's great to hear :-) At the moment, it isn't possible to add these new strategies, because we are refactoring a large part of the code related to...

Progress with https://github.com/meilisearch/milli/pull/601 , but I'm keeping this issue open because I'd like to have more comprehensive snapshot tests still.

Regarding https://github.com/meilisearch/milli/pull/656 , part of the reason that I did not continue developing it (besides time) is that it would increase the amount of memory used during indexing (up to...

I'd like to propose a particular strategy on Meilisearch's side to deal with this. 1. We limit the number of search requests that can be simultaneously executed. This number of...