snowykami/meilisearch
1
0
Fork 0
mirror of https://github.com/meilisearch/meilisearch.git synced 2024-11-25 11:35:05 +08:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

Update README.md (#630)

Browse Source 
* Update README.md

* Update README.md

Co-Authored-By: Clément Renault <renault.cle@gmail.com>

Co-authored-by: Clément Renault <renault.cle@gmail.com>
This commit is contained in:
karan bamal 2020-04-24 13:41:27 +05:30 committed by GitHub
parent 8c41fb2b49
commit 1b7a6687c8
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 33 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
README.md
View File

@ -163,40 +163,8 @@ You can access the web interface in your web browser at the root of the server.
### Documentation
Now that your MeiliSearch server is up and running, you can learn more about how to tune your search engine in [the documentation](https://docs.meilisearch.com).
Now that your MeiliSearch server is up and running, you can learn more about how to tune your search engine in [the documentation](https://docs.meilisearch.com).
### Technical features
- Provides [6 default ranking criteria](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-core/src/criterion/mod.rs#L106-L111) used to [bucket sort](https://en.wikipedia.org/wiki/Bucket_sort) documents
- Accepts [custom criteria](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-core/src/criterion/mod.rs#L20-L29) and can apply them in any custom order
- Supports [ranged queries](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-core/src/query_builder.rs#L342), useful for paginating results
- Can [distinct](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-core/src/query_builder.rs#L324-L329) and [filter](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-core/src/query_builder.rs#L313-L318) returned documents based on context defined rules
- Searches for [concatenated](https://github.com/meilisearch/MeiliSearch/pull/164) and [splitted query words](https://github.com/meilisearch/MeiliSearch/pull/232) to improve the search quality.
- Can store complete documents or only [user schema specified fields](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/datasets/movies/schema.toml)
- The [default tokenizer](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-tokenizer/src/lib.rs) can index Latin based languages and Kanji characters
- Returns [the matching text areas](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-types/src/lib.rs#L49-L65), useful to highlight matched words in results
- Accepts query time search config like the [searchable attributes](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-core/src/query_builder.rs#L331-L336)
- Supports [runtime incremental indexing](https://github.com/meilisearch/MeiliSearch/blob/3ea5aa18a209b6973b921542d46a79e1c753c163/meilisearch-core/src/store/mod.rs#L143-L212)
## Performance
When processing a dataset composed of 5M books, each with their own titles and authors, MeiliSearch is able to carry out more than 553 req/sec with an average response time of 21 ms on an Intel i7-7700 (8) @ 4.2GHz.
Requests are made using [wrk](https://github.com/wg/wrk) and scripted to simulate real users' queries.
```
Running 10s test @ http://1.2.3.4:7700
2 threads and 10 connections
Thread Stats Avg Stdev Max +/- Stdev
Latency 21.45ms 15.64ms 214.10ms 85.95%
Req/Sec 256.48 37.66 330.00 69.50%
5132 requests in 10.05s, 2.31MB read
Requests/sec: 510.46
Transfer/sec: 234.77KB
```
We also indexed a dataset containing about _12 millions_ cities names in _24 minutes_ on a _8 cores_, _64 GB of RAM_, and a _300 GB NMVe_ SSD machine.<br/>
The size of the resulting database reached _16 GB_ and search results were presented between _30 ms_ and _4 seconds_ for short prefix queries.
## Contributing