Spring Boot 如何提升服務(wù)吞吐量？

背景

生產(chǎn)環(huán)境偶爾會有一些慢請求導致系統(tǒng)性能下降，吞吐量下降，下面介紹幾種優(yōu)化建議。

方案

1、undertow替換tomcat

電子商務(wù)類型網(wǎng)站大多都是短請求，一般響應時間都在100ms，這時可以將web容器從tomcat替換為undertow，下面介紹下步驟：?

1、增加pom配置

????org.springframework.boot
????spring-boot-starter-web
????
????????
????????????org.springframework.boot
????????????spring-boot-starter-tomcat
????????
????


????org.springframework.boot
????spring-boot-starter-undertow

2、增加相關(guān)配置

server:
??undertow:
????direct-buffers:?true
????io-threads:?4
????worker-threads:?160

重新啟動可以在控制臺看到容器已經(jīng)切換為undertow了。

推薦閱讀：吊打 Tomcat ，Undertow 性能很炸??！

2、緩存

將部分熱點數(shù)據(jù)或者靜態(tài)數(shù)據(jù)放到本地緩存或者redis中，如果有需要可以定時更新緩存數(shù)據(jù)

3、異步

在代碼過程中我們很多代碼都不需要等返回結(jié)果，也就是部分代碼是可以并行執(zhí)行，這個時候可以使用異步，最簡單的方案是使用springboot提供的@Async注解，當然也可以通過線程池來實現(xiàn)，下面簡單介紹下異步步驟。?

1、pom依賴 一般springboot引入web相關(guān)依賴就行

????org.springframework.boot
????spring-boot-starter-web

2、在啟動類中增加@EnableAsync注解

@EnableAsync
@SpringBootApplicationpublic?
class?AppApplication
{????
????public?static?void?main(String[]?args){????????
??????SpringApplication.run(AppApplication.class,?args);
????}
}

3、需要時在指定方法中增加@Async注解，如果是需要等待返回值，則demo如下

@Async
public?Future?doReturn(int?i){????????
??try?{????????????
??????//?這個方法需要調(diào)用500毫秒
??????Thread.sleep(500);
??}?catch?(InterruptedException?e)?{
??????e.printStackTrace();
??}????????//?消息匯總
??return?new?AsyncResult<>("異步調(diào)用");
}

4、如果有線程變量或者logback中的mdc，可以增加傳遞

import?org.slf4j.MDC;
import?org.springframework.context.annotation.Configuration;
import?org.springframework.core.task.TaskDecorator;
import?org.springframework.scheduling.annotation.AsyncConfigurerSupport;
import?org.springframework.scheduling.annotation.EnableAsync;
import?org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor;

import?java.util.Map;
import?java.util.concurrent.Executor;

/**
?*?@Description:
?*/
@EnableAsync
@Configuration
public?class?AsyncConfig?extends?AsyncConfigurerSupport?{
????@Override
????public?Executor?getAsyncExecutor()?{
????????ThreadPoolTaskExecutor?executor?=?new?ThreadPoolTaskExecutor();
????????executor.setTaskDecorator(new?MdcTaskDecorator());
????????executor.initialize();
????????return?executor;
????}
}

class?MdcTaskDecorator?implements?TaskDecorator?{

????@Override
????public?Runnable?decorate(Runnable?runnable)?{
????????Map?contextMap?=?MDC.getCopyOfContextMap();
????????return?()?->?{
????????????try?{
????????????????MDC.setContextMap(contextMap);
????????????????runnable.run();
????????????}?finally?{
????????????????MDC.clear();
????????????}
????????};
????}
}

5、有時候異步需要增加阻塞

import?lombok.extern.slf4j.Slf4j;
import?org.springframework.context.annotation.Bean;
import?org.springframework.context.annotation.Configuration;
import?org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor;

import?java.util.concurrent.Executor;
import?java.util.concurrent.ThreadPoolExecutor;

@Configuration
@Slf4j
public?class?TaskExecutorConfig?{

????@Bean("localDbThreadPoolTaskExecutor")
????public?Executor?threadPoolTaskExecutor()?{
????????ThreadPoolTaskExecutor?taskExecutor?=?new?ThreadPoolTaskExecutor();
????????taskExecutor.setCorePoolSize(5);
????????taskExecutor.setMaxPoolSize(200);
????????taskExecutor.setQueueCapacity(200);
????????taskExecutor.setKeepAliveSeconds(100);
????????taskExecutor.setThreadNamePrefix("LocalDbTaskThreadPool");
????????taskExecutor.setRejectedExecutionHandler((Runnable?r,?ThreadPoolExecutor?executor)?->?{
????????????????????if?(!executor.isShutdown())?{
????????????????????????try?{
????????????????????????????Thread.sleep(300);
????????????????????????????executor.getQueue().put(r);
????????????????????????}?catch?(InterruptedException?e)?{
????????????????????????????log.error(e.toString(),?e);
????????????????????????????Thread.currentThread().interrupt();
????????????????????????}
????????????????????}
????????????????}
????????);
????????taskExecutor.initialize();
????????return?taskExecutor;
????}


}

4、業(yè)務(wù)拆分

可以將比較耗時或者不同的業(yè)務(wù)拆分出來提供單節(jié)點的吞吐量

5、集成消息隊列

有很多場景對數(shù)據(jù)實時性要求不那么強的，或者對業(yè)務(wù)進行業(yè)務(wù)容錯處理時可以將消息發(fā)送到kafka，然后延時消費。

舉個例子，根據(jù)條件查詢指定用戶發(fā)送推送消息，這里可以時按時、按天、按月等等，這時就