Reduce Join工作原理
Map端的主要工作:对来自不同表或文件的key/value对,打上标签以区别不同来源的记录。然后用连接字段作为key,其余部分和新加标志作为value,最后进行输出
Reduce端的主要工作:在Reduce端以连接字段作为key的分组已经完成,我们只需要在每一个分组中将哪些来源于不同记录分开,最后进行合并。
编程案例
- 创建商品和合并后的Bean类
package com.atguigu.mapreduce.table;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
import org.apache.hadoop.io.Writable;
public class TableBean implements Writable {
private String order_id; // 订单id
private String p_id; // 产品id
private int amount; // 产品数量
private String pname; // 产品名称
private String flag; // 表的标记
public TableBean() {
super();
}
public TableBean(String order_id, String p_id, int amount, String pname, String flag) {
super();
this.order_id = order_id;
this.p_id = p_id;
this.amount = amount;
this.pname = pname;
this.flag = flag;
}
public String getFlag() {
return flag;
}
public void setFlag(String flag) {
this.flag = flag;
}
public String getOrder_id() {
return order_id;
}
public void setOrder_id(String order_id) {
this.order_id = order_id;
}
public String getP_id() {
return p_id;
}
public void setP_id(String p_id) {
this.p_id = p_id;
}
public int getAmount() {
return amount;
}
public void setAmount(int amount) {
this.amount = amount;
}
public String getPname() {
return pname;
}
public void setPname(String pname) {
this.pname = pname;
}
@Override
public void write(DataOutput out) throws IOException {
out.writeUTF(order_id);
out.writeUTF(p_id);
out.writeInt(amount);
out.writeUTF(pname);
out.writeUTF(flag);
}
@Override
public void readFields(DataInput in) throws IOException {
this.order_id = in.readUTF();
this.p_id = in.readUTF();
this.amount = in.readInt();
this.pname = in.readUTF();
this.flag = in.readUTF();
}
@Override
public String toString() {
return order_id + " " + pname + " " + amount + " " ;
}
}
- 编写TableMapper类,获取输入文件名称,键k为连接值,比如两个表的共有属性,输出(k,bean)
package com.atguigu.mapreduce.table;
import java.io.IOException;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.lib.input.FileSplit;
public class TableMapper extends Mapper<LongWritable, Text, Text, TableBean>{
String name;
TableBean bean = new TableBean();
Text k = new Text();
@Override
protected void setup(Context context) throws IOException, InterruptedException {
// 1 获取输入文件切片
FileSplit split = (FileSplit) context.getInputSplit();
// 2 获取输入文件名称
name = split.getPath().getName();
}
@Override
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
// 1 获取输入数据
String line = value.toString();
// 2 不同文件分别处理
if (name.startsWith("order")) {// 订单表处理
// 2.1 切割
String[] fields = line.split(" ");
// 2.2 封装bean对象
bean.setOrder_id(fields[0]);
bean.setP_id(fields[1]);
bean.setAmount(Integer.parseInt(fields[2]));
bean.setPname("");
bean.setFlag("order");
k.set(fields[1]);
}else // 产品表处理
// 2.3 切割
String[] fields = line.split(" ");
// 2.4 封装bean对象
bean.setP_id(fields[0]);
bean.setPname(fields[1]);
bean.setFlag("pd");
bean.setAmount(0);
bean.setOrder_id("");
k.set(fields[0]);
}
// 3 写出
context.write(k, bean);
}
}
- 编写TableReducer类,合并两个表的内容,输出(bean,nullWritable)
package com.atguigu.mapreduce.table;
import java.io.IOException;
import java.util.ArrayList;
import org.apache.commons.beanutils.BeanUtils;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;
public class TableReducer extends Reducer<Text, TableBean, TableBean, NullWritable> {
@Override
protected void reduce(Text key, Iterable<TableBean> values, Context context) throws IOException, InterruptedException {
// 1准备存储订单的集合
ArrayList<TableBean> orderBeans = new ArrayList<>();
// 2 准备bean对象
TableBean pdBean = new TableBean();
for (TableBean bean : values) {
if ("order".equals(bean.getFlag())) {// 订单表
// 拷贝传递过来的每条订单数据到集合中
TableBean orderBean = new TableBean();
try {
BeanUtils.copyProperties(orderBean, bean);
} catch (Exception e) {
e.printStackTrace();
}
orderBeans.add(orderBean);
} else {// 产品表
try {
// 拷贝传递过来的产品表到内存中
BeanUtils.copyProperties(pdBean, bean);
} catch (Exception e) {
e.printStackTrace();
}
}
}
// 3 表的拼接
for(TableBean bean:orderBeans){
bean.setPname (pdBean.getPname());
// 4 数据写出去
context.write(bean, NullWritable.get());
}
}
}
- 编写TableDriver类
package com.atguigu.mapreduce.table;
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 TableDriver {
public static void main(String[] args) throws Exception {
// 0 根据自己电脑路径重新配置
args = new String[]{"e:/input/inputtable","e:/output1"};
// 1 获取配置信息,或者job对象实例
Configuration configuration = new Configuration();
Job job = Job.getInstance(configuration);
// 2 指定本程序的jar包所在的本地路径
job.setJarByClass(TableDriver.class);
// 3 指定本业务job要使用的Mapper/Reducer业务类
job.setMapperClass(TableMapper.class);
job.setReducerClass(TableReducer.class);
// 4 指定Mapper输出数据的kv类型
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(TableBean.class);
// 5 指定最终输出的数据的kv类型
job.setOutputKeyClass(TableBean.class);
job.setOutputValueClass(NullWritable.class);
// 6 指定job的输入原始文件所在目录
FileInputFormat.setInputPaths(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
// 7 将job中配置的相关参数,以及job所用的java类所在的jar包, 提交给yarn去运行
boolean result = job.waitForCompletion(true);
System.exit(result ? 0 : 1);
}
}
Reduce Join的缺点:合并操作在reduce阶段完成,Reduce端处理压力大,而Map端资源利用率不高,易产生数据倾斜。
解决方案:在Map端实现数据合并
Map Join
Map Join适用于一张表十分小,一张表十分大的场景。
用处:在Map端缓存多张表,提前处理业务逻辑,减少Reduce端的压力,减少数据倾斜。
方法:在Mapper的setup阶段,将小表缓存到集合中。而后在map阶段拼接。
编程案例
不需要Reduce阶段,设置ReduceTask数量为0.
- 在驱动模块中添加缓存文件
package test;
import java.net.URI;
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 DistributedCacheDriver {
public static void main(String[] args) throws Exception {
// 0 根据自己电脑路径重新配置
args = new String[]{"e:/input/inputtable2", "e:/output1"};
// 1 获取job信息
Configuration configuration = new Configuration();
Job job = Job.getInstance(configuration);
// 2 设置加载jar包路径
job.setJarByClass(DistributedCacheDriver.class);
// 3 关联map
job.setMapperClass(DistributedCacheMapper.class);
// 4 设置最终输出数据类型
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(NullWritable.class);
// 5 设置输入输出路径
FileInputFormat.setInputPaths(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
// 6 加载缓存数据
job.addCacheFile(new URI("file:///e:/input/inputcache/pd.txt"));
// 7 Map端Join的逻辑不需要Reduce阶段,设置reduceTask数量为0
job.setNumReduceTasks(0);
// 8 提交
boolean result = job.waitForCompletion(true);
System.exit(result ? 0 : 1);
}
}
- 读取缓存的文件数据
package test;
import java.io.BufferedReader;
import java.io.FileInputStream;
import java.io.IOException;
import java.io.InputStreamReader;
import java.util.HashMap;
import java.util.Map;
import org.apache.commons.lang.StringUtils;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
public class DistributedCacheMapper extends Mapper<LongWritable, Text, Text, NullWritable>{
Map<String, String> pdMap = new HashMap<>();
@Override
protected void setup(Mapper<LongWritable, Text, Text, NullWritable>.Context context) throws IOException, InterruptedException {
// 1 获取缓存的文件
URI[] cacheFiles = context.getCacheFiles();
String path = cacheFiles[0].getPath().toString();
BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(path), "UTF-8"));
String line;
while(StringUtils.isNotEmpty(line = reader.readLine())){
// 2 切割
String[] fields = line.split(" ");
// 3 缓存数据到集合
pdMap.put(fields[0], fields[1]);
}
// 4 关流
reader.close();
}
Text k = new Text();
@Override
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
// 1 获取一行
String line = value.toString();
// 2 截取
String[] fields = line.split(" ");
// 3 获取产品id
String pId = fields[1];
// 4 获取商品名称
String pdName = pdMap.get(pId);
// 5 拼接
k.set(line + " "+ pdName);
// 6 写出
context.write(k, NullWritable.get());
}
}