Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/arsenyyankovsky/lucene-serverless
A proof-of-concept serverless full-text search solution built with Apache Lucene and Quarkus framework.
https://github.com/arsenyyankovsky/lucene-serverless
java lucene quarkus serverless
Last synced: 2 days ago
JSON representation
A proof-of-concept serverless full-text search solution built with Apache Lucene and Quarkus framework.
- Host: GitHub
- URL: https://github.com/arsenyyankovsky/lucene-serverless
- Owner: ArsenyYankovsky
- License: mit
- Created: 2021-03-06T16:27:50.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2024-01-09T21:17:38.000Z (9 months ago)
- Last Synced: 2024-09-21T09:04:27.287Z (6 days ago)
- Topics: java, lucene, quarkus, serverless
- Language: Java
- Homepage:
- Size: 118 MB
- Stars: 49
- Watchers: 4
- Forks: 20
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Lucene Serverless
This project demonstrates a proof-of-concept serverless full-text search solution built with Apache Lucene and Quarkus framework.
✔️ No servers
✔️ No fixed costs
✔️ Low (250-300ms) cold starts
⚠️ Better deletion policy is required. Right now old segment files are not deleted as a simple workaround to handle concurrent reads and writes
ℹ️ Cost can be controlled via several factors:
* EFS' [Elastic](https://docs.aws.amazon.com/efs/latest/ug/performance.html#throughput-modes) throughput mode cost more $$, but scales better up and down
* Lambda's [Provisioned Concurrency](https://docs.aws.amazon.com/lambda/latest/dg/provisioned-concurrency.html) provides faster, more consistent first request times but costs more $$
* Lambda memory is kept ~256mb, tune based on index size and observed speeds. Note: vCPU is allocated proportional to memory.
Please note that the project is not ready for production since I haven't tested it under a prolonged load and to be honest interfaces need to be nicer.
Read the blog post about it [here](https://medium.com/@arsenyyankovski/serverless-full-text-search-with-aws-lambda-and-efs-cf24e1b6fe3b)
## Prerequisites
- [Serverless framework >= 1.56.1](https://serverless.com/framework/docs/getting-started/)
- AWS account
## Run it
1. Replace region, vpc id and subnets in the `serverless.yml` file
2. Deploy the stack
`sls deploy`
3. Don't forget to remove it if you're not planning to use it
`sls remove`
### Index a document
URL: `https://.execute-api..amazonaws.com/dev/index`
HTTP method: POST
Example request body:
```json
{
"indexName": "books",
"documents": [
{
"name": "The Foundation",
"author": "Isaac Asimov"
}
]
}
```
### Query documents
URL: `https://.execute-api..amazonaws.com/dev/query`
HTTP method: POST
Example request body:
```json
{
"indexName": "books",
"query": "author:isaac"
}
```
Example response body:
```json
{
"totalDocuments": "1",
"documents": [
{
"author": "Isaac Asimov",
"name": "The Foundation"
}
]
}
```
## Build native image
`./mvnw clean package`