why is meilisearch search data so fast? #1732
Replies: 3 comments
-
Hello @FTLLOVE! Thanks for your interest 🙂 |
Beta Was this translation helpful? Give feedback.
-
Hey @FTLLOVE, Thank you for using MeiliSearch! For instance, we generate:
There is something like eight different databases that are indexed at indexing time and this is why the indexing speed can be a little slow sometimes. By using these inverted indexes we avoid doing a linear ranking system, a system that sorts one by one the documents and then ranks them according to a score, we prefer computing the scores by batches instead and doing big old intersections and unions of ids of documents. Doing this makes big datasets be searchable in a fast way. |
Beta Was this translation helpful? Give feedback.
-
thanks for your answer, i understand some , i will continue to use MeiliSearch as the search engine of the project team. i will recommend it to my friends.i wish that MeiliSearch organization is getting better and better!come on |
Beta Was this translation helpful? Give feedback.
-
In fact, this is not a bug, just curious, i am from China, the company currently use MeiliSearch as search engine,not using ElasticSearch. thank you for MeiliSearch so easy to use a search engine, i am curious why MeiliSearch is so fast, I think not only because of the development language is Rust, So I want to get some knowledge,Thanks
Beta Was this translation helpful? Give feedback.
All reactions