大数据MR模型以及代码实现
数据：
[customers.txt]
1,tom,12
2,tom,13
3,tom,14
4,tom,15 [orders.txt]
1,no001,12.23,1
2,no001,12.23,1
3,no001,12.23,2
4,no001,12.23,2
5,no001,12.23,2
6,no001,12.23,3
7,no001,12.23,3
8,no001,12.23,3
9,no001,12.23,3map端join
1.创建Mapper
package com.it18zhang.hdfs.mr.mapjoin;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.util.HashMap;
import java.util.Map;
/**
* join操作，map端连接。
*/
public class MapJoinMapper extends Mapper {
private Map allCustomers = new HashMap();
//启动,初始化客户信息
protected void setup(Context context) throws IOException, InterruptedException {
try {
Configuration conf = context.getConfiguration();
FileSystem fs = FileSystem.get(conf);
FSDataInputStream fis = fs.open(new Path("file:///d:/mr/mapjoin/customers.txt"));
//得到缓冲区阅读器
BufferedReader br = new BufferedReader(new InputStreamReader(fis));
String line = null ;
while((line = br.readLine()) != null){
//得到cid
String cid = line.substring(0,line.indexOf(","));
allCustomers.put(cid,line);
}
} catch (Exception e) {
e.printStackTrace();
}
}
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
//订单信息
String line = value.toString();
//提取customer id
String cid = line.substring(line.lastIndexOf(",") + 1);
//订单信息
String orderInfo = line.substring(0,line.lastIndexOf(","));
//连接customer + "," + order
String customerInfo = allCustomers.get(cid);
context.write(new Text(customerInfo + "," + orderInfo),NullWritable.get());
}
} 2.创建App
package com.it18zhang.hdfs.mr.mapjoin;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
/**
*
*/
public class MapJoinApp {
public static void main(String[] args) throws Exception {
Configuration conf = new Configuration();
conf.set("fs.defaultFS","file:///");
Job job = Job.getInstance(conf);
//设置job的各种属性
job.setJobName("MapJoinApp"); //作业名称
job.setJarByClass(MapJoinApp.class); //搜索类
//添加输入路径
FileInputFormat.addInputPath(job,new Path(args[0]));
//设置输出路径
FileOutputFormat.setOutputPath(job,new Path(args[1]));
//没有reduce
job.setNumReduceTasks(0);
job.setMapperClass(MapJoinMapper.class); //mapper类
job.setMapOutputKeyClass(Text.class); //
job.setMapOutputValueClass(NullWritable.class); //
job.waitForCompletion(true);
}
}
join端连接
-----------------------
1.自定义key
package com.it18zhang.hdfs.mr.mapjoin.reducejoin;
import org.apache.hadoop.io.WritableComparable;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
/**
*/
public class ComboKey2 implements WritableComparable {
//0-customer 1-order
private int type ;
private int cid ;
private int oid ;
private String customerInfo = "" ;
private String orderInfo = "" ;
public int compareTo(ComboKey2 o) {
int type0 = o.type ;
int cid0= o.cid;
int oid0 = o.oid;
String customerInfo0 = o.customerInfo;
String orderInfo0 = o.orderInfo ;
//是否同一个customer的数据
if(cid == cid0){
//同一个客户的两个订单
if(type == type0){
return oid - oid0 ;
}
//一个Customer + 他的order
else{
if(type ==0)
return -1 ;
else
return 1 ;
}
}
//cid不同
else{
return cid - cid0 ;
}
}
public void write(DataOutput out) throws IOException {
out.writeInt(type);
out.writeInt(cid);
out.writeInt(oid);
out.writeUTF(customerInfo);
out.writeUTF(orderInfo);
}
public void readFields(DataInput in) throws IOException {
this.type = in.readInt();
this.cid = in.readInt();
this.oid = in.readInt();
this.customerInfo = in.readUTF();
this.orderInfo = in.readUTF();
}
} 2.自定义分区类
public class CIDPartitioner extends Partitioner{
public int getPartition(ComboKey2 key, NullWritable nullWritable, int numPartitions) {
return key.getCid() % numPartitions;
}
}
3.创建Mapper
package com.it18zhang.hdfs.mr.mapjoin.reducejoin;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.InputSplit;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.lib.input.FileSplit;
import java.io.IOException;
/**
* mapper
*/
public class ReduceJoinMapper extends Mapper {
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
//
String line = value.toString() ;
//判断是customer还是order
FileSplit split = (FileSplit)context.getInputSplit();
String path = split.getPath().toString();
//客户信息
ComboKey2 key2 = new ComboKey2();
if(path.contains("customers")){
String cid = line.substring(0,line.indexOf(","));
String custInfo = line ;
key2.setType(0);
key2.setCid(Integer.parseInt(cid));
key2.setCustomerInfo(custInfo);
}
//order info
else{
String cid = line.substring(line.lastIndexOf(",") + 1);
String oid = line.substring(0, line.indexOf(","));
String oinfo = line.substring(0, line.lastIndexOf(","));
key2.setType(1);
key2.setCid(Integer.parseInt(cid));
key2.setOid(Integer.parseInt(oid));
key2.setOrderInfo(oinfo);
}
context.write(key2,NullWritable.get());
}
} 4.创建Reducer
package com.it18zhang.hdfs.mr.mapjoin.reducejoin;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;
import java.io.IOException;
import java.util.Iterator;
/**
* ReduceJoinReducer,reducer端连接实现。
*/
public class ReduceJoinReducer extends Reducer
protected void reduce(ComboKey2 key, Iterable
Iterator
it.next();
int type = key.getType();
int cid = key.getCid() ;
String cinfo = key.getCustomerInfo() ;
while(it.hasNext()){
it.next();
String oinfo = key.getOrderInfo();
context.write(new Text(cinfo + "," + oinfo),NullWritable.get());
}
}
}
5.创建排序对比器
package com.it18zhang.hdfs.mr.mapjoin.reducejoin;
import com.it18zhang.hdfs.maxtemp.allsort.secondarysort.ComboKey;
import org.apache.hadoop.io.WritableComparable;
import org.apache.hadoop.io.WritableComparator;
/**
* 组合Key排序对比器
*/
public class ComboKey2Comparator extends WritableComparator {
protected ComboKey2Comparator() {
super(ComboKey2.class, true);
}
public int compare(WritableComparable a, WritableComparable b) {
ComboKey2 k1 = (ComboKey2) a;
ComboKey2 k2 = (ComboKey2) b;
return k1.compareTo(k2);
}
}
6.分组对比器
package com.it18zhang.hdfs.mr.mapjoin.reducejoin;
import com.it18zhang.hdfs.maxtemp.allsort.secondarysort.ComboKey;
import org.apache.hadoop.io.WritableComparable;
import org.apache.hadoop.io.WritableComparator;
/**
* CID分组对比器
*/
public class CIDGroupComparator extends WritableComparator{
protected CIDGroupComparator() {
super(ComboKey2.class, true);
}
public int compare(WritableComparable a, WritableComparable b) {
ComboKey2 k1 = (ComboKey2) a;
ComboKey2 k2 = (ComboKey2) b;
return k1.getCid() - k2.getCid();
}
}
7.App
package com.it18zhang.hdfs.mr.mapjoin.reducejoin;
import com.it18zhang.hdfs.maxtemp.allsort.secondarysort.*;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.input.TextInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
/**
*
*/
public class ReduceJoinApp {
public static void main(String[] args) throws Exception {
Configuration conf = new Configuration();
conf.set("fs.defaultFS","file:///");
Job job = Job.getInstance(conf);
//设置job的各种属性
job.setJobName("ReduceJoinApp"); //作业名称
job.setJarByClass(ReduceJoinApp.class); //搜索类
//添加输入路径
FileInputFormat.addInputPath(job,new Path("D:\\mr\\reducejoin"));
//设置输出路径
FileOutputFormat.setOutputPath(job,new Path("D:\\mr\\reducejoin\\out"));
job.setMapperClass(ReduceJoinMapper.class); //mapper类
job.setReducerClass(ReduceJoinReducer.class); //reducer类
//设置Map输出类型
job.setMapOutputKeyClass(ComboKey2.class); //
job.setMapOutputValueClass(NullWritable.class); //
//设置ReduceOutput类型
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(NullWritable.class); //
//设置分区类
job.setPartitionerClass(CIDPartitioner.class);
//设置分组对比器
job.setGroupingComparatorClass(CIDGroupComparator.class);
//设置排序对比器
job.setSortComparatorClass(ComboKey2Comparator.class);
job.setNumReduceTasks(2); //reduce个数
job.waitForCompletion(true);
}
}
hive
------------------
在hadoop处理结构化数据的数据仓库。
不是: 关系数据库
不是OLTP
实时查询和行级更新。
hive特点
----------
hive存储数据结构(schema)在数据库中,处理的数据进入hdfs.
OLAP
HQL / HiveQL
hive安装
-------------
1.下载hive2.1-tar.gz
2.tar开
$>tar -xzvf hive-2.1.0.tar.gz -C /soft //tar开
$>cd /soft/hive-2.1.0 //
$>ln -s hive-2.1.0 hive //符号连接
3.配置环境变量
[/etc/profile]
HIVE_HOME=/soft/hive
PATH=$PATH$HIVE_HOME/bin
执行 :source /etc/profile
4.验证hive安装成功
$>hive --v
5.配置hive,使用win7的mysql存放hive的元数据.
a)复制mysql驱动程序到hive的lib目录下。
...
b)配置hive-site.xml
复制hive-default.xml.template为hive-site.xml
修改连接信息为mysql链接地址，将${system:...字样替换成具体路径。
[hive/conf/hive-site.xml]
javax.jdo.option.ConnectionPassword
root
password to use against metastore database
javax.jdo.option.ConnectionUserName
root
Username to use against metastore database
javax.jdo.option.ConnectionURL
jdbc:mysql://192.168.231.1:3306/hive2
javax.jdo.option.ConnectionDriverName
com.mysql.jdbc.Driver
Driver class name for a JDBC metastore
6)在msyql中创建存放hive信息的数据库
mysql>create database hive2 ;
6)初始化hive的元数据(表结构)到mysql中。
$>cd /soft/hive/bin
$>schematool -dbType mysql -initSchema
hive命令行操作
------------------------
1.创建hive的数据库
$hive>hive --version //
$hive>hive --help //
$hive>create database mydb2 ; //
$hive>show databases ;
$hive>use mydb2 ;
$hive>create table mydb2.t(id int,name string,age int);
$hive>drop table t ;
$hive>drop table mydb2.t ;
$hive>select * from mydb2.t ; //查看指定库的表
$hive>exit ; //退出
$>hive //hive --service cli
$>hive //hive --service cli
通过远程jdbc方式连接到hive数据仓库
--------------------------------
1.启动hiveserver2服务器，监听端口10000
$>hive --service hiveserver2 &
2.通过beeline命令行连接到hiveserver2
$>beeline //进入beeline命令行(于hive --service beeline)
$beeline>!help //查看帮助
$beeline>!quit //退出
$beeline>!connect jdbc:hive2://localhost:10000/mydb2//连接到hibve数据
$beeline>show databases ;
$beeline>use mydb2 ;
$beeline>show tables; //显式表
使用Hive-jdbc驱动程序采用jdbc方式访问远程数据仓库
----------------------------------------------------
1.创建java模块
2.引入maven
3.添加hive-jdbc依赖
4.0.0
com.it18zhang
HiveDemo
1.0-SNAPSHOT
org.apache.hive
hive-jdbc
2.1.0
4.App
package com.it18zhang.hivedemo;
import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.ResultSet;
import java.sql.Statement;
/**
* 使用jdbc方式连接到hive数据仓库，数据仓库需要开启hiveserver2服务。
*/
public class App {
public static void main(String[] args) throws Exception {
Class.forName("org.apache.hive.jdbc.HiveDriver");
Connection conn = DriverManager.getConnection("jdbc:hive2://192.168.231.201:10000/mydb2");
Statement st = conn.createStatement();
ResultSet rs = st.executeQuery("select id , name ,age from t");
while(rs.next()){
System.out.println(rs.getInt(1) + "," + rs.getString(2)) ;
}
rs.close();
st.close();
conn.close();
}
}hive中表
-------------------
1.managed table
托管表。
删除表时，数据也删除了。
2.external table
外部表。
删除表时，数据不删。
hive命令
----------------
//创建表,external 外部表
$hive>CREATE external TABLE IF NOT EXISTS t2(id int,name string,age int)
COMMENT 'xx' ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' STORED AS TEXTFILE ;
//查看表数据
$hive>desc t2 ;
$hive>desc formatted t2 ;
//加载数据到hive表
$hive>load data local inpath '/home/centos/customers.txt' into table t2 ; //local上传文件
$hive>load data inpath '/user/centos/customers.txt' [overwrite] into table t2 ; //移动文件
//复制表
mysql>create table tt as select * from users ; //携带数据和表结构
mysql>create table tt like users ; //不带数据，只有表结构
hive>create table tt as select * from users ;
hive>create table tt like users ;
//count()查询要转成mr
$hive>select count(*) from t2 ;
$hive>select id,name from t2 ;
//
$hive>select * from t2 order by id desc ; //MR
//启用/禁用表
$hive>ALTER TABLE t2 ENABLE NO_DROP; //不允许删除
$hive>ALTER TABLE t2 DISABLE NO_DROP; //允许删除
//分区表,优化手段之一，从目录的层面控制搜索数据的范围。
//创建分区表.
$hive>CREATE TABLE t3(id int,name string,age int) PARTITIONED BY (Year INT, Month INT) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' ;
//显式表的分区信息
$hive>SHOW PARTITIONS t3;
//添加分区,创建目录
$hive>alter table t3 add partition (year=2014, month=12);
//删除分区
hive>ALTER TABLE employee_partitioned DROP IF EXISTS PARTITION (year=2014, month=11);
//分区结构
hive>/user/hive/warehouse/mydb2.db/t3/year=2014/month=11
hive>/user/hive/warehouse/mydb2.db/t3/year=2014/month=12
//加载数据到分区表
hive>load data local inpath '/home/centos/customers.txt' into table t3 partition(year=2014,month=11);
//创建桶表
$hive>CREATE TABLE t4(id int,name string,age int) CLUSTERED BY (id) INTO 3 BUCKETS ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' ;
//加载数据不会进行分桶操作
$hive>load data local inpath '/home/centos/customers.txt' into table t4 ;
//查询t3表数据插入到t4中。
$hive>insert into t4 select id,name,age from t3 ;
//桶表的数量如何设置?
//评估数据量，保证每个桶的数据量block的2倍大小。
//连接查询
$hive>CREATE TABLE customers(id int,name string,age int) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' ;
$hive>CREATE TABLE orders(id int,orderno string,price float,cid int) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' ;
//加载数据到表
//内连接查询
hive>select a.*,b.* from customers a , orders b where a.id = b.cid ;
//左外
hive>select a.*,b.* from customers a left outer join orders b on a.id = b.cid ;
hive>select a.*,b.* from customers a right outer join orders b on a.id = b.cid ;
hive>select a.*,b.* from customers a full outer join orders b on a.id = b.cid ;
//explode,炸裂,表生成函数。
//使用hive实现单词统计
//1.建表
$hive>CREATE TABLE doc(line string) ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' ;