并发通用库
线程池助手
org.coodex.concurrent.ExecutorsHelper类似于java.util.concurrent.Executors的作用，不同的是：
- 接口提供了定义线程池名称的功能
- 提供了基于任务优先级的线程池，将要执行的任务会按照优先级送入线程池执行，并且，线程优先级也会设置成按照带优先级任务的优先级，高优先级的任务将拿到更多的系统资源
- 扩展了有界阻塞队列，剥离队列大小与线程池最大线程数数的耦合，线程池最大线程数 低于 任务最大值时可用，线程满时，任务队列可以继续容纳一定数量的任务，尝试在线程数和任务数之间达到一个平衡
- TODO: 编写文档的时候想起来的，可以做一个令牌桶的有界队列模型，去峰值
防反跳和连击
前端用得比较多的技术，主要作用是减少频繁发生的事件的处理次数。
可以用于实时性要求不是极高的级频繁数据写入场景。
package org.coodex.concrete.demo.boot;
import org.coodex.concurrent.Coalition;
import org.coodex.concurrent.Debouncer;
import org.coodex.concurrent.Throttler;
import org.coodex.util.Common;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
public class CoalitionDemo {
private final static Logger log = LoggerFactory.getLogger(CoalitionDemo.class);
private static void test(Coalition<Integer> coalition) throws InterruptedException {
for (int i = 0; i < 1000; i++) {
coalition.call(i);
// 随机休眠15毫秒以内
Thread.sleep(Common.random(15));
}
}
public static void main(String[] args) throws InterruptedException {
// 只有空闲15毫秒以后才做一次
test(new Debouncer<>(
arg -> log.info("Debouncer saved: {}", arg), 15
));
// 最多500毫秒执行一次
test(new Throttler<>(
arg -> log.info("Throttler saved: {}", arg), 500
));
}
}
14:34:09.916 [CoalitionPool-2] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 45
14:34:10.069 [CoalitionPool-2] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 66
14:34:10.209 [CoalitionPool-2] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 81
14:34:10.885 [CoalitionPool-1] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 173
14:34:10.905 [CoalitionPool-3] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 174
14:34:11.132 [CoalitionPool-2] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 203
14:34:11.339 [CoalitionPool-1] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 227
14:34:11.358 [CoalitionPool-3] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 230
14:34:11.409 [CoalitionPool-2] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 236
14:34:17.179 [CoalitionPool-2] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Debouncer saved: 999
14:34:17.334 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 0
14:34:17.837 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 61
14:34:18.339 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 118
14:34:18.843 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 178
14:34:19.351 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 245
14:34:19.861 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 317
14:34:19.861 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 318
14:34:20.370 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 385
14:34:20.881 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 453
14:34:21.382 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 529
14:34:21.889 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 606
14:34:22.392 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 675
14:34:22.899 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 756
14:34:23.402 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 815
14:34:23.912 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 885
14:34:24.416 [main] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 947
14:34:25.049 [CoalitionPool-3] INFO org.coodex.concrete.demo.boot.CoalitionDemo - Throttler saved: 999
可以看到，原本需要执行2000次的场景，只进行了20次上下
并行任务
org.coodex.concurrent.Parallel 提供了一个多任务并行处理功能，从根本上来讲，其实就是简化了CountDownLatch的使用。
package org.coodex.concrete.demo.boot;
import com.alibaba.fastjson.JSON;
import org.coodex.concurrent.ExecutorsHelper;
import org.coodex.concurrent.Parallel;
import org.coodex.util.Clock;
import org.coodex.util.Common;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.util.ArrayList;
import java.util.List;
public class ParallelDemo {
private final static Logger log = LoggerFactory.getLogger(ParallelDemo.class);
public static void main(String[] args) {
List<Runnable> runnables = new ArrayList<>();
for (int i = 0; i < 20; i++) {
int finalI = i;
runnables.add(() -> {
try {
Clock.sleep(Common.random(3000));
log.info("task {} finished.", finalI);
} catch (InterruptedException ignore) {
}
});
}
Parallel.Batch batch = new Parallel(
// 5个线程来执行并行任务
ExecutorsHelper.newFixedThreadPool(5, "parallel")
).run(runnables.toArray(new Runnable[0]));
log.info("all task finished.");
log.info("\n{}", JSON.toJSONString(batch, true));
}
}
14:48:34.945 [parallel-3] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 2 finished.
14:48:35.908 [parallel-5] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 4 finished.
14:48:37.242 [parallel-2] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 1 finished.
14:48:37.288 [parallel-4] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 3 finished.
14:48:37.377 [parallel-1] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 0 finished.
14:48:37.398 [parallel-3] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 5 finished.
14:48:37.776 [parallel-3] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 10 finished.
14:48:38.640 [parallel-5] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 6 finished.
14:48:38.827 [parallel-4] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 8 finished.
14:48:39.782 [parallel-4] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 13 finished.
14:48:39.892 [parallel-4] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 14 finished.
14:48:39.910 [parallel-1] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 9 finished.
14:48:39.991 [parallel-2] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 7 finished.
14:48:40.596 [parallel-5] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 12 finished.
14:48:40.719 [parallel-3] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 11 finished.
14:48:41.043 [parallel-2] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 17 finished.
14:48:41.321 [parallel-3] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 19 finished.
14:48:41.478 [parallel-5] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 18 finished.
14:48:42.145 [parallel-4] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 15 finished.
14:48:42.333 [parallel-1] INFO org.coodex.concrete.demo.boot.ParallelDemo - task 16 finished.
14:48:42.333 [main] INFO org.coodex.concrete.demo.boot.ParallelDemo - all task finished.
可以看到，main线程等到所有任务执行完成后才输出all task finished.
{
"end":1565765322333,
"id":"2c963b1966f941e2870fcdd68fdb74f6",
"start":1565765314391,
"tasks":[
{
"end":1565765317377,
"finished":true,
"id":1,
"start":1565765314393,
"timeConsuming":2984
},
{
"end":1565765317242,
"finished":true,
"id":2,
"start":1565765314393,
"timeConsuming":2849
},
{
"end":1565765314945,
"finished":true,
"id":3,
"start":1565765314394,
"timeConsuming":551
},
{
"end":1565765317288,
"finished":true,
"id":4,
"start":1565765314394,
"timeConsuming":2894
},
{
"end":1565765315908,
"finished":true,
"id":5,
"start":1565765314397,
"timeConsuming":1511
},
{
"end":1565765317398,
"finished":true,
"id":6,
"start":1565765314945,
"timeConsuming":2453
},
{
"end":1565765318640,
"finished":true,
"id":7,
"start":1565765315908,
"timeConsuming":2732
},
{
"end":1565765319991,
"finished":true,
"id":8,
"start":1565765317242,
"timeConsuming":2749
},
{
"end":1565765318827,
"finished":true,
"id":9,
"start":1565765317288,
"timeConsuming":1539
},
{
"end":1565765319910,
"finished":true,
"id":10,
"start":1565765317377,
"timeConsuming":2533
},
{
"end":1565765317776,
"finished":true,
"id":11,
"start":1565765317398,
"timeConsuming":378
},
{
"end":1565765320719,
"finished":true,
"id":12,
"start":1565765317776,
"timeConsuming":2943
},
{
"end":1565765320596,
"finished":true,
"id":13,
"start":1565765318640,
"timeConsuming":1956
},
{
"end":1565765319782,
"finished":true,
"id":14,
"start":1565765318827,
"timeConsuming":955
},
{
"end":1565765319892,
"finished":true,
"id":15,
"start":1565765319782,
"timeConsuming":110
},
{
"end":1565765322145,
"finished":true,
"id":16,
"start":1565765319892,
"timeConsuming":2253
},
{
"end":1565765322333,
"finished":true,
"id":17,
"start":1565765319910,
"timeConsuming":2423
},
{
"end":1565765321043,
"finished":true,
"id":18,
"start":1565765319991,
"timeConsuming":1052
},
{
"end":1565765321478,
"finished":true,
"id":19,
"start":1565765320596,
"timeConsuming":882
},
{
"end":1565765321321,
"finished":true,
"id":20,
"start":1565765320719,
"timeConsuming":602
}
],
"timeConsuming":7942
}
返回值对象中，会把每个任务的信息和总览详细返回。