本文重点比较使用虚拟线程的SpringBoot和使用Actix框架的Rust,来实现QR码生成器API。这两种技术都是成熟的,无需进一步介绍。接下来,让我们直接深入测试设置的细节。
一、测试设置
1. 环境
所有测试都在装有16GB RAM的MacBook Pro M1上进行。使用的测试工具是Bombardier的定制版本,支持在请求体中包含随机URL。这些测试使用的软件版本如下:
- SpringBoot 3.1.3,带有Java v20(启用预览以获取虚拟线程)
- Rust 1.72.0
2. 代码
这个QR码生成器应用程序被设计成接收一个JSON请求体,其中包含一个名为"urlToEmbed"的必需参数。该应用程序的主要功能是为指定的URL生成一个QR码,并在HTTP响应中以PNG格式传送QR码。为增加复杂性,该应用程序在HTTPS上运行。
(1) SpringBoot(虚拟线程)
server.port=3000
server.ssl.certificate=/Users/mayankc/Work/source/certs/cert.pem
server.ssl.certificate-private-key=/Users/mayankc/Work/source/certs/key.pem
package com.example.qr;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import org.springframework.boot.web.embedded.tomcat.TomcatProtocolHandlerCustomizer;
import org.springframework.context.annotation.Bean;
import java.util.concurrent.Executors;
@SpringBootApplication
public class QrApplication {
public static void main(String[] args) {
SpringApplication.run(QrApplication.class, args);
}
@Bean
public TomcatProtocolHandlerCustomizer<?> protocolHandlerVirtualThreadExecutorCustomizer() {
return protocolHandler -> {
protocolHandler.setExecutor(Executors.newVirtualThreadPerTaskExecutor());
};
}
}
package com.example.qr;
import org.springframework.web.bind.annotation.PostMapping;
import org.springframework.web.bind.annotation.RequestBody;
import org.springframework.http.ResponseEntity;
import org.springframework.http.HttpStatus;
import org.springframework.http.HttpHeaders;
import org.springframework.web.bind.annotation.RestController;
import java.util.Optional;
import com.example.qr.QrRequest;
import com.example.qr.QrGenerator;
@RestController
public class QrController {
@PostMapping("/qr")
public ResponseEntity handleRequest(@RequestBody QrRequest qrRequest) {
if(qrRequest.getUrlToEmbed() == null) {
return new ResponseEntity<>(HttpStatus.BAD_REQUEST);
}
try {
HttpHeaders httpHeaders = new HttpHeaders();
httpHeaders.add(HttpHeaders.CONTENT_TYPE, "image/png");
return new ResponseEntity<byte[]>(
QrGenerator.generateQR(qrRequest.getUrlToEmbed(), 512, 512),
httpHeaders,
HttpStatus.OK);
} catch (Exception e) {
return new ResponseEntity<>(HttpStatus.INTERNAL_SERVER_ERROR);
}
}
}
package com.example.qr;
import java.io.ByteArrayOutputStream;
import java.io.IOException;
import com.google.zxing.BarcodeFormat;
import com.google.zxing.WriterException;
import com.google.zxing.client.j2se.MatrixToImageConfig;
import com.google.zxing.client.j2se.MatrixToImageWriter;
import com.google.zxing.common.BitMatrix;
import com.google.zxing.qrcode.QRCodeWriter;
public class QrGenerator {
public static byte[] generateQR(String text, int width, int height) throws WriterException, IOException {
QRCodeWriter qrCodeWriter = new QRCodeWriter();
BitMatrix bitMatrix = qrCodeWriter.encode(text, BarcodeFormat.QR_CODE, width, height);
ByteArrayOutputStream pngOutputStream = new ByteArrayOutputStream();
MatrixToImageConfig con = new MatrixToImageConfig() ;
MatrixToImageWriter.writeToStream(bitMatrix, "PNG", pngOutputStream, con);
byte[] pngData = pngOutputStream.toByteArray();
return pngData;
}
}
package com.example.qr;
public class QrRequest {
private String urlToEmbed;
public String getUrlToEmbed() {
return this.urlToEmbed;
}
public void setUrlToEmbed(String urlToEmbed) {
this.urlToEmbed = urlToEmbed;
}
}
(2) Rust
[package]
name = "actix_qr_generator"
version = "0.1.0"
edition = "2021"
[dependencies]
actix-web = { version = "4", features = ["openssl"] }
qrcode-generator = "4.1.8"
serde = { version = "1.0", features = ["derive"] }
serde_json = "1"
openssl = { version = "0.10" , features = ["vendored"] }
use actix_web::{web, post, App, HttpServer, HttpResponse, Responder};
use qrcode_generator::QrCodeEcc;
use serde::Deserialize;
use openssl::ssl::{SslAcceptor, SslFiletype, SslMethod};
#[derive(Deserialize)]
struct QrRequest {
urlToEmbed: String,
}
#[post("/qr")]
async fn generate_qr(qr_request: web::Json<QrRequest>) -> impl Responder {
if qr_request.urlToEmbed.is_empty() {
return HttpResponse::BadRequest().into();
}
let result: Vec<u8> = qrcode_generator::to_png_to_vec(qr_request.urlToEmbed.clone(), QrCodeEcc::Low, 512)
.unwrap();
return HttpResponse::Ok()
.content_type("image/png")
.body(result);
}
#[actix_web::main]
async fn main() -> std::io::Result<()> {
let mut builder = SslAcceptor::mozilla_intermediate(SslMethod::tls()).unwrap();
builder
.set_private_key_file("/Users/mayankc/Work/source/perfComparisons/certs/key.pem", SslFiletype::PEM)
.unwrap();
builder
.set_certificate_chain_file("/Users/mayankc/Work/source/perfComparisons/certs/cert.pem")
.unwrap();
HttpServer::new(|| App::new().service(generate_qr))
.bind_openssl("127.0.0.1:3000", builder)?
.run()
.await
}
// 注意 ================================================
// 该应用程序已在发布模式下构建。
// =====================================================
二、结果
为了全面评估性能,这里进行了一系列细致的检查。每个检查包括10万个请求,并在10、50和100个并发连接的范围内评估它们的效率。考虑到QR码生成的资源密集型特性,故意保持了稍低的请求量,与其他场景相比。
结果如下:
根据以下公式,还生成了一个得分卡。对于每个测量,获取获胜的差距。如果获胜的差距是:
- < 5%,不给予任何分数
- 在5%到20%之间,获胜者得1分
- 在20%到50%之间,获胜者得2分
- 50%,获胜者得3分
得分卡如下: