浅析MySQL读写分离与负载均衡实例教程

MySQL 数据库的读写分离和负载均衡一般是通过第三方软件来实现的,也可以通过mysql驱动程序来实现,如com.mysql.jdbc.ReplicationDriver.

详细文档参见:http://dev.mysql.com/doc/refman/5.5/en/connector-j-info.html,代码如下:

1. importjava.sql.Connection;
2. importjava.sql.ResultSet;
3. importjava.util.Properties;
5. importcom.mysql.jdbc.ReplicationDriver;
7. publicclassReplicationDriverDemo{
9. publicstaticvoidmain(String[]args)throwsException{
10. ReplicationDriverdriver=newReplicationDriver();
12. Propertiesprops=newProperties();
14. //Wewantthisforfailoverontheslaves
15. props.put("autoReconnect","true");
17. //Wewanttoloadbalancebetweentheslaves
18. props.put("roundRobinLoadBalance","true");
20. props.put("user","foo");
21. props.put("password","bar");
23. //
24. //LookslikeanormalMySQLJDBCurl,witha
25. //comma-separatedlistofhosts,thefirst
26. //beingthe'master',therestbeinganynumber
27. //ofslavesthatthedriverwillloadbalanceagainst
28. //
30. Connectionconn=
31. driver.connect("jdbc:mysql:replication://master,slave1,slave2,slave3/test",
32. props);
34. //
35. //Performread/writeworkonthemaster
36. //bysettingtheread-onlyflagto"false"
37. //
39. conn.setReadOnly(false);
40. conn.setAutoCommit(false);
41. conn.createStatement().executeUpdate("UPDATEsome_table....");
42. conn.commit();
44. //
45. //Now,doaqueryfromaslave,thedriverautomaticallypicksone
46. //fromthelist--phpfensi.com
47. //
49. conn.setReadOnly(true);
51. ResultSetrs=
52. conn.createStatement().executeQuery("SELECTa,bFROMalt_table");
54. .......
55. }
56. }

读写分离:

1. jdbc:mysql:replication://master:3306,slave1:3306,slave2:3306/dbname
3. Whenusingthefollowingconnectionstring:jdbc:mysql:replication://dbmaster:3306,dbslave1:3306,dbslave2:3306/dbname
5. dbmasterisusedforallwriteconnectionsasexpectedanddbslave1isusedforallreadconnections,butdbslave2isneverused.Iwouldhaveexpecteddistributedreadsbetweendbslave1anddbslave2.

原理是:ReplicationDriver生成代理的connection对象,当设置这个connection.readOnly=true时,连接slave,当connection.readOnly=false时,连接master

负载均衡:

1. jdbc:mysql:loadbalance://master:3306,slave1:3306,slave2:3306/dbname
3. Whenusingthefollowingconnectionstring:jdbc:mysql:loadbalance://dbmaster:3306,dbslave1:3306,dbslave2:3306/dbname
4. connectionsareload-balancedbetweenallthreeserversforbothreadandwriteconnections.

问题:读写分离时可能会碰到刚写完master,再马上到slave进行查询的情况,而主从复制的时候有延迟,这时怎么解决呢？有两个办法.

1.比如增加页面保存数据后马上跳转到列表页面,这时可能出不来数据,因为复制还没完成，这时可以在前台添加一些成功的提示,成功页面等进行一些页面跳转延迟处理,让服务器有时间去复制,复制延迟一般在毫秒级,而这种提示处理在秒级,所以时间上一般是足够的.

2.第1种办法可能部分场景是可行的,但是有些场景要求比较高,需要实时的,这时可以在读取的时候进行处理,强制从master中读取,可以通过注解,加参数/标识等来指定从master读取数据.