How to implement a Reactive Restful API service based on webflux and Netty?

Question

I want to implement a RESTful API service based on Webflux and Netty, and I have found some implementation demos on the official website and the Internet, but according to the current implementation method. I tested it on Jmeter and it doesn't seem to work very well, I wonder if it's a problem with my implementation, or if it's a problem with my thinking。

application.yml

Server:
  port:10001
spring:
  r2dbc:
    url:r2dbc:mssql://localhost:1433/TutorialDB
    username:sa
    password:123456
    pool:
      max-size:32
      max-idle-time:2m
      max-life-time:10m
      acquire-retry:3

pom.xml

    <dependency>
      <groupId>org.springframework.boot</groupId>
      <artifactId>spring-boot-starter</artifactId>
    </dependency>
    <dependency>
      <groupId>org.springframework.boot</groupId>
      <artifactId>spring-boot-starter-webflux</artifactId>
      <version>3.1.2</version> 
    </dependency>
    <dependency>
      <groupId>io.projectreactor.netty</groupId>
      <artifactId>reactor-netty</artifactId>
    </dependency>
    <dependency>
      <groupId>io.projectreactor</groupId>
      <artifactId>reactor-core</artifactId>
      <version>3.5.8</version>
    </dependency>
    <dependency>
      <groupId>io.projectreactor</groupId>
      <artifactId>reactor-test</artifactId>
      <scope>test</scope>
      <version>3.6.2</version>
    </dependency>
    <dependency>
      <groupId>org.springframework.boot</groupId>
      <artifactId>spring-boot-starter-test</artifactId>
      <version>3.1.2</version>
    </dependency>
    <dependency>
      <groupId>io.projectreactor</groupId>
      <artifactId>reactor-test</artifactId>
    </dependency>
    <dependency>
      <groupId>io.projectreactor</groupId>
      <artifactId>reactor-core</artifactId>
      <version>3.6.2</version>
      <scope>compile</scope>
    </dependency>
    <dependency>
      <groupId>org.slf4j</groupId>
      <artifactId>slf4j-api</artifactId>
    </dependency>
    <dependency>
      <groupId>org.springframework.data</groupId>
      <artifactId>spring-data-r2dbc</artifactId>
      <version>3.1.2</version>
    </dependency>
    <dependency>
      <groupId>io.r2dbc</groupId>
      <artifactId>r2dbc-pool</artifactId>
      <version>1.0.0.RELEASE</version>
    </dependency>
    <dependency>
      <groupId>io.r2dbc</groupId>
      <artifactId>r2dbc-mssql</artifactId>
    </dependency>
    <dependency>
      <groupId>com.mysql</groupId>
      <artifactId>mysql-connector-j</artifactId>
      <scope>runtime</scope>
    </dependency>

    <dependency>
      <groupId>io.asyncer</groupId>
      <artifactId>r2dbc-mysql</artifactId>
      <version>1.0.6</version>
    </dependency>

Repository

public interface DeviceTypeDetailRepository extends ReactiveCrudRepository<DeviceTypeDetail, Long>{ 
}

Service

@Service
public class DeviceTypeDetailServiceImpl implements DeviceTypeDetailService {

    private final DeviceTypeDetailRepository deviceTypeDetailRepository;
    
    public DeviceTypeDetailServiceImpl(DeviceTypeDetailRepository deviceTypeDetailRepository) {
        this.deviceTypeDetailRepository = deviceTypeDetailRepository;
    }
    @Override
    public Flux<DeviceTypeDetail> FetchAll(){
        return deviceTypeDetailRepository.findAll();
    }

}

Controller

@CrossOrigin
@RestController
@RequestMapping("/api/devicetypedetail")
public class DeviceTypeDetailController {
@Autowired
private DeviceTypeDetailService deviceTypeDetailService;

    @GetMapping("/all")
    public Flux<DeviceTypeDetail> FetchAll() {
        return deviceTypeDetailService.FetchAll();
    }

}

Jmeter Aggregate Report --Webflux
Number of Threads = 1000 Ramp-up = 1 loop Count = 100

Jmeter Aggregate Report --MVC
Number of Threads = 1000 Ramp-up = 1 loop Count = 100

I wonder if it's the wrong way I've implemented it or what's the case.

Answer 1

Never mind, I give up on using the combination of WebFlux + R2DBC. It's possible that R2DBC is causing a decrease in throughput compared to MVC + JDBC. When I use Vert.x alone, it feels good, and there's a throughput improvement of 3-5 times.