数据结构
customer表
| 1 |
hanmeimei |
ShangHai |
110 |
| 2 |
leilei |
BeiJing |
112 |
| 3 |
lucy |
GuangZhou |
119 |
oder表
| 1 | 1 | 50 |
| 2 | 1 | 200 |
| 3 | 3 | 15 |
| 4 | 3 | 350 |
| 5 | 3 | 58 |
| 6 | 1 | 42 |
| 7 | 1 | 352 |
| 8 | 2 | 1135 |
| 9 | 2 | 400 |
| 10 | 2 | 2000 |
| 11 | 2 | 300 |
MAPJOIN
场景：我们模拟一个有一份小表一个大表的场景，customer是那份小表，order是那份大表
做法：直接将较小的数据加载到内存中，按照连接的关键字建立索引，
大份数据作为MapTask的输入键值对 map()方法的每次输入都去内存当中直接去匹配连接。
然后把连接结果按 key 输出，这种方法要使用 hadoop中的 DistributedCache 把小份数据分布到各个计算节点，
每个 maptask 执行任务的节点都需要加载该数据到内存，并且按连接关键字建立索引。
环境配置：因为我们是在本地操作的，所以需要配置本地的hadoop
1。下载hadoop
2.解压到一个目录，记住，一会要用
配置电脑环境变量
如果你是一个初学者那么你就创建一个Java工程，步骤自己搜吧，网上一大堆然后创建一个mapjoin的包，在包里创建一个
JoinDemo的类然后如下第24行代码，在JoinDemo后加extened往后就根据我的代码敲就好了。
package mapjoin;##以上内容标志了段代码是在mapjoin这个包里 import org.apache.hadoop.fs.FileSystem; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.conf.Configured; import org.apache.hadoop.fs.FSDataInputStream; import org.apache.hadoop.fs.Path; import org.apache.hadoop.io.LongWritable; import org.apache.hadoop.io.Text; import org.apache.hadoop.io.WritableComparable; import org.apache.hadoop.mapreduce.Job; import org.apache.hadoop.mapreduce.Mapper; import org.apache.hadoop.mapreduce.Reducer; import org.apache.hadoop.mapreduce.lib.input.FileInputFormat; import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat; import org.apache.hadoop.util.Tool; import org.apache.hadoop.util.ToolRunner; import java.io.*; import java.net.URI; import java.util.HashMap; import java.util.Map; ##以上内容是需要导入的jar包，否则一些代码会报红。
1 public class JoinDemo extends Configured implements Tool {
2 // customer文件在本地上的位置。
3 // TODO: 改用参数传入
4 private static final String CUSTOMER_CACHE_URL = "input/customer.txt"; --本地输入文件路径
5 // "hdfs://hadoop1:9000/user/hadoop/mapreduce/cache/customer.txt";
6 private static class CustomerBean { --
7 private int custId;
8 private String name;
9 private String address;
10 private String phone;
11
12 public CustomerBean() {}
13
14 public CustomerBean(int custId, String name, String address,String phone) {
16 super();
17 this.custId = custId;
18 this.name = name;
19 this.address = address;
20 this.phone = phone;
21 }
22
23
24
25 public int getCustId() {
26 return custId;
27 }
28
29 public String getName() {
30 return name;
31 }
32
33 public String getAddress() {
34 return address;
35 }
36
37 public String getPhone() {
38 return phone;
39 }
40 }
41
42 private static class CustOrderMapOutKey implements
43 WritableComparable<CustOrderMapOutKey> {
44 private int custId;
45 private int orderId;
46
47 public void set(int custId, int orderId) {
48 this.custId = custId;
49 this.orderId = orderId;
50 }
51
52 public int getCustId() {
53 return custId;
54 }
55
56 public int getOrderId() {
57 return orderId;
58 }
59
60
61 public void write( DataOutput out) throws IOException {
62 out.writeInt(custId);
63 out.writeInt(orderId);
64 }
65
66
67 public void readFields( DataInput in) throws IOException {
68 custId = in.readInt();
69 orderId = in.readInt();
70 }
71
72 public int compareTo(CustOrderMapOutKey o) {
73 int res = custId -o.custId;
74 return res == 0 ? orderId-o.orderId : res;
75 }
76
77 public boolean equals(Object obj) {
78 if (obj instanceof CustOrderMapOutKey) {
79 CustOrderMapOutKey o = (CustOrderMapOutKey)obj;
80 return custId == o.custId && orderId == o.orderId;
81 } else {
82 return false;
83 }
84 }
85
86 @Override
87 public String toString() {
88 return custId + "\t" + orderId;
89 }
90 }
91
92 private static class JoinMapper extends
93 Mapper<LongWritable, Text, CustOrderMapOutKey, Text> {
94 private final CustOrderMapOutKey outputKey = new CustOrderMapOutKey();
95 private final Text outputValue = new Text();
96
97
98 /**
99 * 在内存中customer数据
100 */
101 private static final Map<Integer, CustomerBean> CUSTOMER_MAP = new HashMap<Integer, CustomerBean>();
102
103 @Override
104 protected void setup(Context context)
105 throws IOException, InterruptedException {
106 //读取缓存的文件
107 FileSystem fs = FileSystem
108 .get( URI.create(CUSTOMER_CACHE_URL), context.getConfiguration());
109 FSDataInputStream fdis = fs.open(new Path(CUSTOMER_CACHE_URL));
110
111 BufferedReader reader = new BufferedReader(new InputStreamReader(fdis));
112 String line = null;
113 String[] cols = null;
114
115 // 格式：客户编号 姓名 地址 电话
116 while ((line = reader.readLine()) != null) {
117 cols = line.split("\t");
118 if (cols.length < 4) { // 数据格式不匹配，忽略
119 continue;
120 }
121
122 CustomerBean bean = new CustomerBean(Integer.parseInt(cols[0]), cols[1], cols[2], cols[3]);
123 //缓存客户信息
124 CUSTOMER_MAP.put(bean.getCustId(), bean);
125 }
126 }
127
128 @Override
129 protected void map(LongWritable key, Text value, Mapper.Context context)
130 throws IOException, InterruptedException {
131
132 // 格式: 订单编号 客户编号 订单金额
133 String[] cols = value.toString().split("\t");
134 if (cols.length < 3) {
135 return;
136 }
137
138 int custId = Integer.parseInt(cols[1]); // 取出客户编号
139 CustomerBean customerBean = CUSTOMER_MAP.get(custId);
140
141 if (customerBean == null) { // 没有对应的customer信息可以连接
142 return;
143 }
144
145 StringBuffer sb = new StringBuffer();
146 sb.append(cols[2])
147 .append("\t")
148 .append(customerBean.getName())
149 .append("\t")
150 .append(customerBean.getAddress())
151 .append("\t")
152 .append(customerBean.getPhone());
153
154 outputValue.set(sb.toString());
155 outputKey.set(custId, Integer.parseInt(cols[0]));
156
157 //WC 操作
158 // context.write(value, 1);
159
160 context.write(outputKey, outputValue);
161 }
162
163 }
164
165 /**
166 * reduce
167 * @author Ivan
168 *
169 */
170 private static class JoinReducer extends
171 Reducer<CustOrderMapOutKey, Text, CustOrderMapOutKey, Text> {
172 @Override
173 protected void reduce(CustOrderMapOutKey key, Iterable<Text> values, Context context)
174 throws IOException, InterruptedException {
175 // 什么事都不用做，直接输出
176 for (Text value : values) {
177 context.write(key, value);
178 }
179 }
180 }
181 /**
182 * @param args
183 * @throws Exception
184 */
185 public static void main(String[] args) throws Exception {
186 // if (args.length < 2) {
187 // new IllegalArgumentException("Usage: <inpath> <outpath>");
188 // return;
189 // }
190 //调用 JoinDemo 对象的run方法
191 ToolRunner.run(new Configuration(), new JoinDemo(), args);
192 }
193
194
195 public int run(String[] args) throws Exception {
196 System.setProperty("hadoop.home.dir", "D:\\software\\hadoop-2.6.0\\hadoop-2.6.0");
197 //默认配置
198 Configuration conf = getConf();
199 FileSystem fileSystem=FileSystem.get(conf);
200 Path outputPath=new Path("output/");
201 if(fileSystem.exists(outputPath)){
202 fileSystem.delete(outputPath,true);
203 }
204
205 Job job = Job.getInstance(conf);
206 job.setJarByClass(JoinDemo.class);
207
208 // 添加customer cache文件
209 job.addCacheFile(URI.create(CUSTOMER_CACHE_URL));
210 Path inputPath=new Path("input/order.txt");
211 // Path inputPath= new Path(args[0])
212 FileInputFormat.addInputPath(job, inputPath);
213
214 // Path outputPath=new Path(args[1]);
215 FileOutputFormat.setOutputPath(job, outputPath);
216
217 // map settings
218 job.setMapperClass(JoinMapper.class);
219 job.setMapOutputKeyClass(CustOrderMapOutKey.class);
220 job.setMapOutputValueClass(Text.class);
221
222 // reduce settings
223 // job.setReducerClass(JoinReducer.class);
224 // job.setOutputKeyClass(CustOrderMapOutKey.class);
225 // job.setOutputKeyClass(Text.class);
226 boolean res = job.waitForCompletion(true);
227 return res ? 0 : 1;
228 }
229 }
本地的文件位置
最后执行结果：
原文地址：https://www.cnblogs.com/xuziyu/p/10803264.html
时间： 2024-08-30 11:48:23