并发框架执行多任务的小实践

并发框架定义

public class FutureTaskWorker {
    # 查看处理器数
    private static final int processors = Runtime.getRuntime().availableProcessors();

    # 定义线程池
    private static final ExecutorService executor = Executors.newFixedThreadPool(processors + 1);

    static {
        # 查看一下机器处理器数
        System.out.println(processors + " processors");
    }

	# 实现方式1----------------------------------------------------------------------------
    public static <T, R> List<R> handle(List<T> tasks, Function<T, R> function) {
        List<CompletableFuture<R>> featureRes = new ArrayList<>();
        for (T task : tasks) {
            # 任务添加到线程池
            CompletableFuture<R> feature = CompletableFuture.supplyAsync(() -> function.apply(task), executor);
            featureRes.add(feature);
        }

        List<R> res = new ArrayList<>();
        try {
            # 等所有任务执行完成
            CompletableFuture.allOf(featureRes.toArray(new CompletableFuture[0])).get();

            # 收集执行结果
            for (CompletableFuture<R> feature : featureRes) {
                res.add(feature.get());
            }
            return res;
        } catch (InterruptedException | ExecutionException e) {
            throw new RuntimeException(e);
        }
    }
    
	# 实现方式2----------------------------------------------------------------------------
    public static <T, R> List<R> handle1(List<T> tasks, Function<T, R> function) {
    	# 任务添加到线程池
        List<CompletableFuture<R>> futureList = tasks.stream().map(task -> CompletableFuture.supplyAsync(() -> function.apply(task), executor)).collect(Collectors.toList());
        
        # 等所有任务执行完成
        CompletableFuture<Void> allCompletableFuture = CompletableFuture.allOf(futureList.toArray(new CompletableFuture[0]));
        
        # 收集执行结果
        return allCompletableFuture.thenApply(e -> futureList.stream().map(CompletableFuture::join).collect(Collectors.toList())).join();
    }
    
    public static void shunDown() {
        executor.shutdownNow();
    }
}

并发框架使用

public static void main(String[] args) {
    List<String> names = new ArrayList<>();
    names.add("小明1");
    names.add("小明2");
    names.add("小明3");
    names.add("小明4");
    names.add("小明5");
    names.add("小明6");
    names.add("小明7");
    names.add("小明8");
    names.add("小明9");
    
    try {
    	# 使用并发框架来执行多任务
        List<String> res = handle1(names, name -> businessMethod(name));
        
        res.forEach(System.out::println);
    } finally {
    	# 关闭线程池
        shunDown();
    }
}

# 这里模拟业务方法
public static String businessMethod(String str) {
    System.out.println(Thread.currentThread().getName() + "线程:" + new Date());
    return str + "  " + new Date();
}

# 执行结果
16 processors
pool-1-thread-1线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-9线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-6线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-5线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-3线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-2线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-7线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-8线程:Fri Nov 03 15:18:10 CST 2023
pool-1-thread-4线程:Fri Nov 03 15:18:10 CST 2023
小明1  Fri Nov 03 15:18:10 CST 2023
小明2  Fri Nov 03 15:18:10 CST 2023
小明3  Fri Nov 03 15:18:10 CST 2023
小明4  Fri Nov 03 15:18:10 CST 2023
小明5  Fri Nov 03 15:18:10 CST 2023
小明6  Fri Nov 03 15:18:10 CST 2023
小明7  Fri Nov 03 15:18:10 CST 2023
小明8  Fri Nov 03 15:18:10 CST 2023
小明9  Fri Nov 03 15:18:10 CST 2023

参考:

使用CompletableFuture.allOf实现异步执行同步搜集结果
CompletableFuture用法详解

你可能感兴趣的:(并发编程,最佳实践,windows)