serhii-samoilenko / hsa-homework-11

Elasticsearch: Index & Autocomplete

Highload Software Architecture 8 Lesson 11 Homework

Index & Autocomplete

Test project setup

The demo is written in Kotlin/Quarkus and uses docker-compose to run Elasticsearch.

The com/example/Demo.kt/runDemo function is used to set up index and run autocomplete queries. It also produces a report in a Markdown format. The final report is located in the REPORT.md file.

Please refer to the REPORT.md file for the information on mapping configuration and queries.

After launching the demo, you can open simple UI on http://localhost:8080 to try the queries yourself.

Issues

I couldn't find a way to make Elasticsearch to leverage exactly 3 typos while performing suggestion search. The best I've managed to achieve is the empirically found coefficients of "minimum_should_match": "7<30% 10<60%" on the search query, which gives relatively stable results for words longer than 9 characters.

Words shorter than 9 characters are mostly handled by the fuzzy suggestion search, which handles up to 2 typos.

How to build and run

Start up Elasticsearch

docker-compose up -d

Build and run demo application (Requires Java 17+)

./gradlew build && \
java -jar build/quarkus-app/quarkus-run.jar

You can also run application in dev mode that enables live coding using:

./gradlew quarkusDev