import java.io.File
import java.io.FileInputStream
import java.io.InputStreamReader
import java.io.PrintWriter
import java.net.HttpURLConnection
import java.net.URL
import java.util.HashMap
import java.util.Map
import java.util.concurrent.ExecutorService
import java.util.concurrent.Executors
import java.util.concurrent.Semaphore
public class ConcurrentTest {
private static int thread_num = 200
private static int client_num = 460
private static Map keywordMap = new HashMap()
static {
try {
InputStreamReader isr = new InputStreamReader(new FileInputStream(
new File("clicks.txt")), "GBK")
BufferedReader buffer = new BufferedReader(isr)
String line = ""
while ((line = buffer.readLine()) != null) {
keywordMap.put(line.substring(0, line.lastIndexOf(":")), "")
}
} catch (Exception e) {
e.printStackTrace()
}
}
public static void main(String[] args) {
int size = keywordMap.size()
// TODO Auto-generated method stub
ExecutorService exec = Executors.newCachedThreadPool()
// 50个线程可以同时访问
final Semaphore semp = new Semaphore(thread_num)
// 模拟2000个客户端访问
for (int index = 0 index < client_num index++) {
final int NO = index
Runnable run = new Runnable() {
public void run() {
try {
// 获取许可
semp.acquire()
System.out.println("Thread:" + NO)
String host = "http://10.99.23.42:7001/KMQueryCenter/query.do?"
String para = "method=getQueryResult&pageNum=1&pageSize=5&"
+ "queryKeyWord="
+ getRandomSearchKey(NO)
+ "&questionID=-1&questionIdPath=-1&searchType=1"
+ "&proLine=&proSeries=&proType=" + NO
System.out.println(host + para)
URL url = new URL(host)// 此处填写供测试的url
HttpURLConnection connection = (HttpURLConnection) url
.openConnection()
// connection.setRequestMethod("POST")
// connection.setRequestProperty("Proxy-Connection",
// "Keep-Alive")
connection.setDoOutput(true)
connection.setDoInput(true)
PrintWriter out = new PrintWriter(connection
.getOutputStream())
out.print(para)
out.flush()
out.close()
BufferedReader in = new BufferedReader(
new InputStreamReader(connection
.getInputStream()))
String line = ""
String result = ""
while ((line = in.readLine()) != null) {
result += line
}
// System.out.println(result)
// Thread.sleep((long) (Math.random()) * 1000)
// 释放
System.out.println("第：" + NO + " 个")
semp.release()
} catch (Exception e) {
e.printStackTrace()
}
}
}
exec.execute(run)
}
// 退出线程池
exec.shutdown()
}
private static String getRandomSearchKey(final int no) {
String ret = ""
int size = keywordMap.size()
// int wanna = (int) (Math.random()) * (size - 1)
ret = (keywordMap.entrySet().toArray())[no].toString()
ret = ret.substring(0, ret.lastIndexOf("="))
System.out.println("\t" + ret)
return ret
}
}
思路：压力测试原理有两种，一种是连接测试，一种是负载测试
1，连接测试，就是多线程，不停的请求直到服务器死机或达到预期效果即可！
2，负载测试，就是一个线程的多步操作，每个线程占用服务器cpu 内存是否符合节能高效的标准，如果不是，那就优化吧
建议用LOADRUNNER吧。
建议别老盯着测试结果,如果出现这种情况，看一看程序日志和数据库的日志，尤其是数据库SQL日志，看看每个SQL完成的时间变化，应该是耗时越来越长，这样就要优化了。建议先搞数据库压力测试，根据情况制作10万-200万测试记录，目标是优化SQL语句，大中型项目中，这时第一项。
接下来再搞逻辑层和Web层的压力测试，目标是优化业务逻辑实现算法。
