如何合理地估算线程池大小?
这个问题虽然看起来很小,却并不那么容易回答。大家如果有更好的方法欢迎赐教,先来一个天真的估算方法:假设要求一个系统的TPS(Transaction Per Second或者Task Per Second)至少为20,然后假设每个Transaction由一个线程完成,继续假设平均每个线程处理一个Transaction的时间为4s。那么问题转化为:如何设计线程池大小,使得可以在1s内处理完20个Transaction?
计算过程很简单,每个线程的处理能力为0.25TPS,那么要达到20TPS,显然需要20/0.25=80个线程。
很显然这个估算方法很天真,因为它没有考虑到CPU数目。一般服务器的CPU核数为16或者32,如果有80个线程,那么肯定会带来太多不必要的线程上下文切换开销。
再来第二种简单的但不知是否可行的方法(N为CPU总核数):
如果是CPU密集型应用,则线程池大小设置为N+1
如果是IO密集型应用,则线程池大小设置为2N+1
如果一台服务器上只部署这一个应用并且只有这一个线程池,那么这种估算或许合理,具体还需自行测试验证。
接下来在这个文档:服务器性能IO优化 中发现一个估算公式:
最佳线程数目 = ((线程等待时间+线程CPU时间)/线程CPU时间 )* CPU数目
比如平均每个线程CPU运行时间为0.5s,而线程等待时间(非CPU运行时间,比如IO)为1.5s,CPU核心数为8,那么根据上面这个公式估算得到:((0.5+1.5)/0.5)*8=32。这个公式进一步转化为:
最佳线程数目 = (线程等待时间与线程CPU时间之比 + 1)* CPU数目
可以得出一个结论:线程等待时间所占比例越高,需要越多线程。线程CPU时间所占比例越高,需要越少线程。
上一种估算方法也和这个结论相合。
一个系统最快的部分是CPU,所以决定一个系统吞吐量上限的是CPU。增强CPU处理能力,可以提高系统吞吐量上限。但根据短板效应,真实的系统吞吐量并不能单纯根据CPU来计算。那要提高系统吞吐量,就需要从“系统短板”(比如网络延迟、IO)着手:
尽量提高短板操作的并行化比率,比如多线程下载技术
增强短板能力,比如用NIO替代IO
第一条可以联系到Amdahl定律,这条定律定义了串行系统并行化后的加速比计算公式:
加速比=优化前系统耗时 / 优化后系统耗时
加速比越大,表明系统并行化的优化效果越好。Addahl定律还给出了系统并行度、CPU数目和加速比的关系,加速比为Speedup,系统串行化比率(指串行执行代码所占比率)为F,CPU数目为N:
Speedup <= 1 / (F + (1-F)/N)
当N足够大时,串行化比率F越小,加速比Speedup越大。
写到这里,我突然冒出一个问题。
是否使用线程池就一定比使用单线程高效呢?
答案是否定的,比如Redis就是单线程的,但它却非常高效,基本操作都能达到十万量级/s。从线程这个角度来看,部分原因在于:
多线程带来线程上下文切换开销,单线程就没有这种开销
锁
当然“Redis很快”更本质的原因在于:Redis基本都是内存操作,这种情况下单线程可以很高效地利用CPU。而多线程适用场景一般是:存在相当比例的IO和网络操作。
所以即使有上面的简单估算方法,也许看似合理,但实际上也未必合理,都需要结合系统真实情况(比如是IO密集型或者是CPU密集型或者是纯内存操作)和硬件环境(CPU、内存、硬盘读写速度、网络状况等)来不断尝试达到一个符合实际的合理估算值。
最后来一个“Dark Magic”估算方法(因为我暂时还没有搞懂它的原理),使用下面的类:
1 packagethreadpool;2
3 importjava.math.BigDecimal;4 importjava.math.RoundingMode;5 importjava.util.Timer;6 importjava.util.TimerTask;7 importjava.util.concurrent.BlockingQueue;8
9 /**
10 * A class that calculates the optimal thread pool boundaries. It takes the11 * desired target utilization and the desired work queue memory consumption as12 * input and retuns thread count and work queue capacity.13 *14 *@authorNiklas Schlimm15 */
16 public abstract classPoolSizeCalculator {17
18 /**
19 * The sample queue size to calculate the size of a single {@linkRunnable}20 * element.21 */
22 private final int SAMPLE_QUEUE_SIZE = 1000;23
24 /**
25 * Accuracy of test run. It must finish within 20ms of the testTime26 * otherwise we retry the test. This could be configurable.27 */
28 private final int EPSYLON = 20;29
30 /**
31 * Control variable for the CPU time investigation.32 */
33 private volatile booleanexpired;34
35 /**
36 * Time (millis) of the test run in the CPU time calculation.37 */
38 private final long testtime = 3000;39
40 /**
41 * Calculates the boundaries of a thread pool for a given {@linkRunnable}.42 *43 *@paramtargetUtilization the desired utilization of the CPUs (0 <= targetUtilization <= * 1) *@paramtargetQueueSizeBytes * the desired maximum work queue size of the thread pool (bytes)44 */
45 protected voidcalculateBoundaries(BigDecimal targetUtilization, BigDecimal targetQueueSizeBytes) {46 calculateOptimalCapacity(targetQueueSizeBytes);47 Runnable task =creatTask();48 start(task);49 start(task); //warm up phase
50 long cputime =getCurrentThreadCPUTime();51 start(task); //test intervall
52 cputime = getCurrentThreadCPUTime() -cputime;53 long waittime = (testtime * 1000000) -cputime;54 calculateOptimalThreadCount(cputime, waittime, targetUtilization);55 }56
57 private voidcalculateOptimalCapacity(BigDecimal targetQueueSizeBytes) {58 long mem =calculateMemoryUsage();59 BigDecimal queueCapacity = targetQueueSizeBytes.divide(newBigDecimal(mem),60 RoundingMode.HALF_UP);61 System.out.println("Target queue memory usage (bytes): "
62 +targetQueueSizeBytes);63 System.out.println("createTask() produced " + creatTask().getClass().getName() + " which took " + mem + " bytes in a queue");64 System.out.println("Formula: " + targetQueueSizeBytes + " / " +mem);65 System.out.println("* Recommended queue capacity (bytes): " +queueCapacity);66 }67
68 /**
69 * Brian Goetz' optimal thread count formula, see 'Java Concurrency in70 * * Practice' (chapter 8.2) *71 * *@paramcpu72 * * cpu time consumed by considered task73 * *@paramwait74 * * wait time of considered task75 * *@paramtargetUtilization76 * * target utilization of the system77 */
78 private void calculateOptimalThreadCount(long cpu, longwait,79 BigDecimal targetUtilization) {80 BigDecimal waitTime = newBigDecimal(wait);81 BigDecimal computeTime = newBigDecimal(cpu);82 BigDecimal numberOfCPU = newBigDecimal(Runtime.getRuntime()83 .availableProcessors());84 BigDecimal optimalthreadcount =numberOfCPU.multiply(targetUtilization)85 .multiply(new BigDecimal(1).add(waitTime.divide(computeTime,86 RoundingMode.HALF_UP)));87 System.out.println("Number of CPU: " +numberOfCPU);88 System.out.println("Target utilization: " +targetUtilization);89 System.out.println("Elapsed time (nanos): " + (testtime * 1000000));90 System.out.println("Compute time (nanos): " +cpu);91 System.out.println("Wait time (nanos): " +wait);92 System.out.println("Formula: " + numberOfCPU + " * "
93 + targetUtilization + " * (1 + " + waitTime + " / "
94 + computeTime + ")");95 System.out.println("* Optimal thread count: " +optimalthreadcount);96 }97
98 /**
99 * * Runs the {@linkRunnable} over a period defined in {@link#testtime}.100 * * Based on Heinz Kabbutz' ideas101 * * (http://www.javaspecialists.eu/archive/Issue124.html).102 * *103 * *@paramtask104 * * the runnable under investigation105 */
106 public voidstart(Runnable task) {107 long start = 0;108 int runs = 0;109 do{110 if (++runs > 5) {111 throw new IllegalStateException("Test not accurate");112 }113 expired = false;114 start =System.currentTimeMillis();115 Timer timer = newTimer();116 timer.schedule(newTimerTask() {117 public voidrun() {118 expired = true;119 }120 }, testtime);121 while (!expired) {122 task.run();123 }124 start = System.currentTimeMillis() -start;125 timer.cancel();126 } while (Math.abs(start - testtime) >EPSYLON);127 collectGarbage(3);128 }129
130 private void collectGarbage(inttimes) {131 for (int i = 0; i < times; i++) {132 System.gc();133 try{134 Thread.sleep(10);135 } catch(InterruptedException e) {136 Thread.currentThread().interrupt();137 break;138 }139 }140 }141
142 /**
143 * Calculates the memory usage of a single element in a work queue. Based on144 * Heinz Kabbutz' ideas145 * (http://www.javaspecialists.eu/archive/Issue029.html).146 *147 *@returnmemory usage of a single {@linkRunnable} element in the thread148 * pools work queue149 */
150 public longcalculateMemoryUsage() {151 BlockingQueue queue =createWorkQueue();152 for (int i = 0; i < SAMPLE_QUEUE_SIZE; i++) {153 queue.add(creatTask());154 }155
156 long mem0 = Runtime.getRuntime().totalMemory() -Runtime.getRuntime().freeMemory();157 long mem1 = Runtime.getRuntime().totalMemory() -Runtime.getRuntime().freeMemory();158
159 queue = null;160
161 collectGarbage(15);162
163 mem0 = Runtime.getRuntime().totalMemory() -Runtime.getRuntime().freeMemory();164 queue =createWorkQueue();165
166 for (int i = 0; i < SAMPLE_QUEUE_SIZE; i++) {167 queue.add(creatTask());168 }169
170 collectGarbage(15);171
172 mem1 = Runtime.getRuntime().totalMemory() -Runtime.getRuntime().freeMemory();173
174 return (mem1 - mem0) /SAMPLE_QUEUE_SIZE;175 }176
177 /**
178 * Create your runnable task here.179 *180 *@returnan instance of your runnable task under investigation181 */
182 protected abstractRunnable creatTask();183
184 /**
185 * Return an instance of the queue used in the thread pool.186 *187 *@returnqueue instance188 */
189 protected abstractBlockingQueue createWorkQueue();190
191 /**
192 * Calculate current cpu time. Various frameworks may be used here,193 * depending on the operating system in use. (e.g.194 *http://www.hyperic.com/products/sigar). The more accurate the CPU time195 * measurement, the more accurate the results for thread count boundaries.196 *197 *@returncurrent cpu time of current thread198 */
199 protected abstract longgetCurrentThreadCPUTime();200
201 }
然后自己继承这个抽象类并实现它的三个抽象方法,比如下面是我写的一个示例(任务是请求网络数据),其中我指定期望CPU利用率为1.0(即100%),任务队列总大小不超过100,000字节:
1 packagethreadpool;2
3 importjava.io.BufferedReader;4 importjava.io.IOException;5 importjava.io.InputStreamReader;6 importjava.lang.management.ManagementFactory;7 importjava.math.BigDecimal;8 importjava.net.HttpURLConnection;9 importjava.net.URL;10 importjava.util.concurrent.BlockingQueue;11 importjava.util.concurrent.LinkedBlockingQueue;12
13 public class SimplePoolSizeCaculatorImpl extendsPoolSizeCalculator {14
15 @Override16 protectedRunnable creatTask() {17 return newAsyncIOTask();18 }19
20 @Override21 protectedBlockingQueue createWorkQueue() {22 return new LinkedBlockingQueue(1000);23 }24
25 @Override26 protected longgetCurrentThreadCPUTime() {27 returnManagementFactory.getThreadMXBean().getCurrentThreadCpuTime();28 }29
30 public static voidmain(String[] args) {31 PoolSizeCalculator poolSizeCalculator = newSimplePoolSizeCaculatorImpl();32 poolSizeCalculator.calculateBoundaries(new BigDecimal(1.0), new BigDecimal(100000));33 }34
35 }36
37 /**
38 * 自定义的异步IO任务39 *@authorWill40 *41 */
42 class AsyncIOTask implementsRunnable {43
44 public voidrun() {45 HttpURLConnection connection = null;46 BufferedReader reader = null;47 try{48 String getURL = "http://baidu.com";49 URL getUrl = newURL(getURL);50
51 connection =(HttpURLConnection) getUrl.openConnection();52 connection.connect();53 reader = new BufferedReader(newInputStreamReader(54 connection.getInputStream()));55
56 String line;57 while ((line = reader.readLine()) != null) {58 //empty loop
59 }60 }61
62 catch(IOException e) {63
64 } finally{65 if(reader != null) {66 try{67 reader.close();68 }69 catch(Exception e) {70
71 }72 }73 connection.disconnect();74 }75
76 }77
78 }
得到如下输出:
Target queue memory usage (bytes): 100000createTask() produced threadpool.AsyncIOTask which took40bytes in a queue
Formula:100000 / 40
* Recommended queue capacity (bytes): 2500Number of CPU:8Target utilization:1Elapsed time (nanos):3000000000Compute time (nanos):280801800Wait time (nanos):2719198200Formula:8 * 1 * (1 + 2719198200 / 280801800)* Optimal thread count: 88
推荐的任务队列大小为2500,线程数为88。依次为依据,我们就可以构造这样一个线程池:
ThreadPoolExecutor pool = new ThreadPoolExecutor(88, 88, 0L, TimeUnit.MILLISECONDS, new LinkedBlockingQueue(2500));
可以将这个文件打包成可执行的jar文件,这样就可以拷贝到测试/正式环境上执行。
1
2 xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
3 4.0.0
4
5 threadpool
6 dark-magic
7 1.0-SNAPSHOT
8 jar
9
10 dark_magic
11 http://maven.apache.org
12
13
14 UTF-8
15
16
17
18
19
20
21
22 dark-magic
23
24
25
26 maven-assembly-plugin
27
28 false
29
30 jar-with-dependencies
31
32
33
34
35 threadpool.SimplePoolSizeCaculatorImpl
36
37
38
39
40
41 make-assembly
42 package
43
44 assembly
45
46
47
48
49
50
51
转载:
http://ifeve.com/how-to-calculate-threadpool-size/
http://www.importnew.com/17384.html
https://www.cnblogs.com/cherish010/p/8334952.html