在网上有一种短8位UUID生成的方法,代码来源:
JAVA生成短8位UUID
public static String[] chars = new String[] { "a", "b", "c", "d", "e", "f",
"g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s",
"t", "u", "v", "w", "x", "y", "z", "0", "1", "2", "3", "4", "5",
"6", "7", "8", "9", "A", "B", "C", "D", "E", "F", "G", "H", "I",
"J", "K", "L", "M", "N", "O", "P", "Q", "R", "S", "T", "U", "V",
"W", "X", "Y", "Z" };
public static String generateShortUuid() {
StringBuffer shortBuffer = new StringBuffer();
String uuid = UUID.randomUUID().toString().replace("-", "");
for (int i = 0; i < 8; i++) {
String str = uuid.substring(i * 4, i * 4 + 4);
int x = Integer.parseInt(str, 16);
shortBuffer.append(chars[x % 0x3E]);
}
return shortBuffer.toString();
}
我们进行测试看到底多少会出现重复,写了一个比较简单的方法:
设置了线程池,数据库连接池,每一个线程进行处理一百万条数据,每次携带7万条数据进行数据库的插入。我们将ID设置为数据库的主键,如果出现错误,则表示数据库ID出现重复现象。
如果需要一次性插入更多的数据,或者在插入的时候报下面的错误:
Packet for query is too large (4,800,048 > 4,194,304)
修改 my.ini 加上 max_allowed_packet =67108864
67108864=64M
默认大小4194304 也就是4M
修改完成之后要重启mysql服务,如果通过命令行修改就不用重启mysql服务。
命令修改:
设置为500M
mysql> set global max_allowed_packet = 500*1024*1024;
查看mysql的max_allowed_packet大小,运行
show VARIABLES like '%max_allowed_packet%';
下面是插入数据代码,进行测试:
private static ThreadPoolExecutor threadPoolExecutor=new ThreadPoolExecutor(50, 100,
0L, TimeUnit.MILLISECONDS,
new LinkedBlockingQueue<Runnable>(1024),new DefaultManagedAwareThreadFactory(), new ThreadPoolExecutor.AbortPolicy());
@Override
public void saveTest() {
threadPoolExecutor.execute(new saveUUID());
}
/**
* 设置保存信息的线程
*/
class saveUUID implements Runnable{
public saveUUID(){
}
@Override
public void run() {
System.out.println("===================>>>>>" + Thread.currentThread().getName());
List<Map<String,Object>> userList = new ArrayList<>();
int index=0;
for(int i = 0; i< 1000000;i++){
long startTime = System.currentTimeMillis();
Map<String,Object> map = new HashMap<>();
map.put("id",getUUID8());
userList.add(map);
index++;
if (index==70000) {
int count = testMapper.saveTest(userList);
if (count > 0) {
userList.clear();
index=0;
long time =System.currentTimeMillis()-startTime;
System.out.println( Thread.currentThread().getName() + "======>>>>" +time );
}
}
}
if(!userList.isEmpty()){
testMapper.saveTest(userList);
}
}
}
XML文件:
<insert id="saveTest" parameterType="java.util.List">
insert into tb_user(id,user_name)
values
<foreach item="item" index="index" collection="list" separator=",">
(#{item.id},#{item.id})
</foreach>
</insert>
进行测试当数据达到:2129580,也就是两百万的时候出现第一次重复,进行数据库数据查询,发现重复原因是因为MySQL不区分大小写。
关于MySQL主键不区分大小写,或则其他查询不区分大小写 Duplicate entry 'AOVbrXXF' for key 'PRIMARY'
将MySQL进行大小写区分。再进行测试。
数据达到400万条无重复,系统插入缓慢。
在基础数据为0的情况下,插入数据100万。系统花费时间为:
26.318秒
在基础数据为0的情况下,插入数据200万。系统花费的时间为:
64.097秒
在基础数据为0的情况下,插入数据,系统花费时间:
DefaultManagedAwareThreadFactory-1======>>>>3209
DefaultManagedAwareThreadFactory-1======>>>>1033
DefaultManagedAwareThreadFactory-1======>>>>1089
DefaultManagedAwareThreadFactory-1======>>>>952
DefaultManagedAwareThreadFactory-1======>>>>921
DefaultManagedAwareThreadFactory-1======>>>>1119
DefaultManagedAwareThreadFactory-1======>>>>825
DefaultManagedAwareThreadFactory-1======>>>>911
DefaultManagedAwareThreadFactory-1======>>>>919
DefaultManagedAwareThreadFactory-1======>>>>850
DefaultManagedAwareThreadFactory-1======>>>>949
DefaultManagedAwareThreadFactory-1======>>>>983
DefaultManagedAwareThreadFactory-1======>>>>720
DefaultManagedAwareThreadFactory-1======>>>>1097
DefaultManagedAwareThreadFactory-1======>>>>862
DefaultManagedAwareThreadFactory-1======>>>>1173
DefaultManagedAwareThreadFactory-1======>>>>942
DefaultManagedAwareThreadFactory-1======>>>>808
DefaultManagedAwareThreadFactory-1======>>>>1159
DefaultManagedAwareThreadFactory-1======>>>>845
DefaultManagedAwareThreadFactory-1======>>>>998
DefaultManagedAwareThreadFactory-1======>>>>1478
DefaultManagedAwareThreadFactory-1======>>>>1420
DefaultManagedAwareThreadFactory-1======>>>>974
DefaultManagedAwareThreadFactory-1======>>>>1398
DefaultManagedAwareThreadFactory-1======>>>>4005
DefaultManagedAwareThreadFactory-1======>>>>2922
DefaultManagedAwareThreadFactory-1======>>>>2126
DefaultManagedAwareThreadFactory-1======>>>>1198
DefaultManagedAwareThreadFactory-1======>>>>931
DefaultManagedAwareThreadFactory-1======>>>>2197
DefaultManagedAwareThreadFactory-1======>>>>1014
DefaultManagedAwareThreadFactory-1======>>>>1231
DefaultManagedAwareThreadFactory-1======>>>>3416
DefaultManagedAwareThreadFactory-1======>>>>4471
DefaultManagedAwareThreadFactory-1======>>>>960
DefaultManagedAwareThreadFactory-1======>>>>1199
DefaultManagedAwareThreadFactory-1======>>>>1142
DefaultManagedAwareThreadFactory-1======>>>>940
DefaultManagedAwareThreadFactory-1======>>>>2211
DefaultManagedAwareThreadFactory-1======>>>>23994
DefaultManagedAwareThreadFactory-1======>>>>37751
DefaultManagedAwareThreadFactory-1======>>>>51163
DefaultManagedAwareThreadFactory-1======>>>>66993
DefaultManagedAwareThreadFactory-1======>>>>79591
DefaultManagedAwareThreadFactory-1======>>>>91846
由上可见当数据量达到一定量的时候,时间成指数上升。
如何能突破这个时间节点,估计还需要进一步优化。
生产UUID满足千万条数据
目前1147万条数据,没有出现重复,满足原作者说的千万条数据不重复