使用JDBC連接MySQL數(shù)據(jù)庫(kù)進(jìn)行數(shù)據(jù)插入的時(shí)候,特別是大批量數(shù)據(jù)連續(xù)插入(10W+),如何提高效率呢?
在JDBC編程接口中Statement 有兩個(gè)方法特別值得注意:
void addBatch() throws SQLException
Adds a set of parameters to this PreparedStatement object's batch of commands.
int[] executeBatch() throws SQLException
Submits a batch of commands to the database for execution and if all commands execute successfully, returns an array of update counts. The int elements of the array that is returned are ordered to correspond to the commands in the batch, which are ordered according to the order in which they were added to the batch.
通過(guò)使用addBatch()和executeBatch()這一對(duì)方法可以實(shí)現(xiàn)批量處理數(shù)據(jù)。
不過(guò)值得注意的是,首先需要在數(shù)據(jù)庫(kù)鏈接中設(shè)置手動(dòng)提交,connection.setAutoCommit(false),然后在執(zhí)行Statement之后執(zhí)行connection.commit()。
package cyl.demo.ipsearcher;
import java.io.BufferedReader;
import java.io.FileInputStream;
import java.io.IOException;
import java.io.InputStreamReader;
import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.PreparedStatement;
import java.sql.SQLException;
public class DbStoreHelper {
private String insert_sql;
private String charset;
private boolean debug;
private String connectStr;
private String username;
private String password;
public DbStoreHelper() {
connectStr = "jdbc:mysql://localhost:3306/db_ip";
// connectStr += "?useServerPrepStmts=falserewriteBatchedStatements=true";
insert_sql = "INSERT INTO tb_ipinfos (iplong1,iplong2,ipstr1,ipstr2,ipdesc) VALUES (?,?,?,?,?)";
charset = "gbk";
debug = true;
username = "root";
password = "***";
}
public void storeToDb(String srcFile) throws IOException {
BufferedReader bfr = new BufferedReader(new InputStreamReader(new FileInputStream(srcFile), charset));
try {
doStore(bfr);
} catch (Exception e) {
e.printStackTrace();
} finally {
bfr.close();
}
}
private void doStore(BufferedReader bfr) throws ClassNotFoundException, SQLException, IOException {
Class.forName("com.mysql.jdbc.Driver");
Connection conn = DriverManager.getConnection(connectStr, username,password);
conn.setAutoCommit(false); // 設(shè)置手動(dòng)提交
int count = 0;
PreparedStatement psts = conn.prepareStatement(insert_sql);
String line = null;
while (null != (line = bfr.readLine())) {
String[] infos = line.split(";");
if (infos.length 5) continue;
if (debug) {
System.out.println(line);
}
psts.setLong(1, Long.valueOf(infos[0]));
psts.setLong(2, Long.valueOf(infos[1]));
psts.setString(3, infos[2]);
psts.setString(4, infos[3]);
psts.setString(5, infos[4]);
psts.addBatch(); // 加入批量處理
count++;
}
psts.executeBatch(); // 執(zhí)行批量處理
conn.commit(); // 提交
System.out.println("All down : " + count);
conn.close();
}
}
執(zhí)行完成以后:
All down : 103498
Convert finished.
All spend time/s : 47
一共10W+,執(zhí)行時(shí)間一共花費(fèi) 47 秒.
這個(gè)效率仍然不高,似乎沒(méi)有達(dá)到想要的效果,需要進(jìn)一步改進(jìn)。
在MySQL JDBC連接字符串中還可以加入?yún)?shù),
rewriteBatchedStatements=true,mysql默認(rèn)關(guān)閉了batch處理,通過(guò)此參數(shù)進(jìn)行打開(kāi),這個(gè)參數(shù)可以重寫(xiě)向數(shù)據(jù)庫(kù)提交的SQL語(yǔ)句。
useServerPrepStmts=false,如果不開(kāi)啟(useServerPrepStmts=false),使用com.mysql.jdbc.PreparedStatement進(jìn)行本地SQL拼裝,最后送到db上就是已經(jīng)替換了?后的最終SQL.
在此稍加改進(jìn),連接字符串中加入下面語(yǔ)句(代碼構(gòu)造方法中去掉注釋):
connectStr += "?useServerPrepStmts=falserewriteBatchedStatements=true";
再次執(zhí)行如下:
All down : 103498
Convert finished.
All spend time/s : 10
同樣的數(shù)據(jù)量,這次執(zhí)行只花費(fèi)了10秒 ,處理效率大大提高.
以上就是本文的全部?jī)?nèi)容,希望對(duì)大家的學(xué)習(xí)有所幫助,也希望大家多多支持腳本之家。
您可能感興趣的文章:- JDBC連接MySQL數(shù)據(jù)庫(kù)批量插入數(shù)據(jù)過(guò)程詳解
- JDBC連接MySql數(shù)據(jù)庫(kù)步驟 以及查詢、插入、刪除、更新等
- java中JDBC實(shí)現(xiàn)往MySQL插入百萬(wàn)級(jí)數(shù)據(jù)的實(shí)例代碼
- Java使用JDBC向MySQL數(shù)據(jù)庫(kù)批次插入10W條數(shù)據(jù)(測(cè)試效率)
- Java使用JDBC或MyBatis框架向Oracle中插入XMLType數(shù)據(jù)
- java實(shí)現(xiàn)jdbc批量插入數(shù)據(jù)
- JDBC鏈接mysql插入數(shù)據(jù)后顯示問(wèn)號(hào)的原因及解決辦法
- JDBC插入數(shù)據(jù)返回?cái)?shù)據(jù)主鍵代碼實(shí)例