本来是看了问答区伙伴的for比线程池快，想也复现一次，然后发现for确实是快些，但是两种方式计算的结果多次运行都不对，在LongAdder的结果倒是对的。

public class LongAccumulatorDemo_6 {
    public static void main(String[] args) throws InterruptedException {
        LongAccumulator accumulator = new LongAccumulator(getFunction(), 1);
        for (int i = 0; i < 5; i++) {
            accMethod(accumulator);
            foriMethod(accumulator);
            Thread.sleep(500);
        }
    }

    private static void foriMethod(LongAccumulator accumulator) {
        long s = System.currentTimeMillis();
        for (int i = 0; i < 10000; i++) {
            accumulator.accumulate(i);
        }
        long e = System.currentTimeMillis();
        System.out.println("fori耗时：" + (e - s));
        System.out.println(accumulator.getThenReset());
    }

    private static void accMethod(LongAccumulator accumulator) {
        ExecutorService service = Executors.newFixedThreadPool(8);
        long start = System.currentTimeMillis();
        IntStream.range(0, 10000).forEach(i -> service.execute(() -> accumulator.accumulate(i)));
        service.shutdown();
        while (!service.isTerminated()) {

        }
        long end = System.currentTimeMillis();
        System.out.println("LongAccumulator耗时：" + (end - start));
        System.out.println(accumulator.getThenReset());
    }

    private static LongBinaryOperator getFunction() {
        return (x, y) -> x + y;
//        return (x, y) -> x * y;
//        return (x, y) -> Math.max(x,y);
    }
}

输出

LongAccumulator耗时：23
49995001
fori耗时：1
49995003
LongAccumulator耗时：14
49995003
fori耗时：0
49995003
LongAccumulator耗时：7
49995003
fori耗时：1
49995003
LongAccumulator耗时：7
49995003
fori耗时：1
49995003
LongAccumulator耗时：8
49995003
fori耗时：0
49995005

以及在

https://coding.imooc.com/learn/questiondetail/Z85pz6GAjQaYn0rV.html

中，您说用LongAccumulator查询数据库，会发现线程池速度明显比for循环快。用线程池查数据库我见过，用LongAccumulator+线程池查数据库有例子吗？