目录
1、先导知识
2、案例
2.1 需求
2.2 代码实现
FlowBean类
Mapper类
Reducer类
Driver类
3、总结
TreeMap底层是根据红黑树的数据结构构建的,默认是根据key的自然排序来组织(比如integer的大小,String的字典排序),如果key是自定义类,可以通过重写compareTo方法自定义排序。
firstKey ()方法 用于返回此TreeMap中具有最小键值的第一个键元素。.
lastKey ()方法 用于返回此TreeMap中具有最大键值的最后一个键元素。.
setup()与cleanup()方法:
1、setup(),此方法被MapReduce框架仅且执行一次,在执行Map任务前,进行相关变量或者资源的集中初始化工作。若是将资源初始化工作放在方法map()中,导致Mapper任务在解析每一行输入时都会进行资源初始化工作,导致重复,程序运行效率不高!
2、cleanup(),此方法被MapReduce框架仅且执行一次,在执行完毕Map任务后,进行相关变量或资源的释放工作。若是将释放资源工作放入方法map()中,也会导致Mapper任务在解析、处理每一行文本后释放资源,而且在下一行文本解析前还要重复初始化,导致反复重复,程序运行效率不高!
这里就使用了cleanup方法,map方法和reduce方法保持TreeMap只有n个元素;cleanup用于输出TreeMap的元素给下一个环节用,只需要执行一次,就放在cleanup。
package com.atguigu.mr.top;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
import org.apache.hadoop.io.WritableComparable;
public class FlowBean implements WritableComparable{
private long upFlow;
private long downFlow;
private long sumFlow;
public FlowBean() {
super();
}
public FlowBean(long upFlow, long downFlow) {
super();
this.upFlow = upFlow;
this.downFlow = downFlow;
}
@Override
public void write(DataOutput out) throws IOException {
out.writeLong(upFlow);
out.writeLong(downFlow);
out.writeLong(sumFlow);
}
@Override
public void readFields(DataInput in) throws IOException {
upFlow = in.readLong();
downFlow = in.readLong();
sumFlow = in.readLong();
}
public long getUpFlow() {
return upFlow;
}
public void setUpFlow(long upFlow) {
this.upFlow = upFlow;
}
public long getDownFlow() {
return downFlow;
}
public void setDownFlow(long downFlow) {
this.downFlow = downFlow;
}
public long getSumFlow() {
return sumFlow;
}
public void setSumFlow(long sumFlow) {
this.sumFlow = sumFlow;
}
@Override
public String toString() {
return upFlow + "\t" + downFlow + "\t" + sumFlow;
}
public void set(long downFlow2, long upFlow2) {
downFlow = downFlow2;
upFlow = upFlow2;
sumFlow = downFlow2 + upFlow2;
}
@Override
public int compareTo(FlowBean bean) {
int result;
if (this.sumFlow > bean.getSumFlow()) {
result = -1;
}else if (this.sumFlow < bean.getSumFlow()) {
result = 1;
}else {
result = 0;
}
return result;
}
}
package com.atguigu.mr.top;
import java.io.IOException;
import java.util.Iterator;
import java.util.TreeMap;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Mapper;
public class TopNMapper extends Mapper{
// 定义一个TreeMap作为存储数据的容器(天然按key排序)
private TreeMap flowMap = new TreeMap();
private FlowBean kBean;
@Override
protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
kBean = new FlowBean();
Text v = new Text();
// 1 获取一行
String line = value.toString();
// 2 切割
String[] fields = line.split("\t");
// 3 封装数据
String phoneNum = fields[0];
long upFlow = Long.parseLong(fields[1]);
long downFlow = Long.parseLong(fields[2]);
long sumFlow = Long.parseLong(fields[3]);
kBean.setDownFlow(downFlow);
kBean.setUpFlow(upFlow);
kBean.setSumFlow(sumFlow);
v.set(phoneNum);
// 4 向TreeMap中添加数据
flowMap.put(kBean, v);
// 5 限制TreeMap的数据量,超过10条就删除掉流量最小的一条数据
if (flowMap.size() > 10) {
// flowMap.remove(flowMap.firstKey());
flowMap.remove(flowMap.lastKey());
}
}
@Override
protected void cleanup(Context context) throws IOException, InterruptedException {
// 6 遍历treeMap集合,输出数据
Iterator bean = flowMap.keySet().iterator();
while (bean.hasNext()) {
FlowBean k = bean.next();
context.write(k, flowMap.get(k));
}
}
}
package com.atguigu.mr.top;
import java.io.IOException;
import java.util.Iterator;
import java.util.TreeMap;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Reducer;
public class TopNReducer extends Reducer {
// 定义一个TreeMap作为存储数据的容器(天然按key排序)
TreeMap flowMap = new TreeMap();
@Override
protected void reduce(FlowBean key, Iterable values, Context context)throws IOException, InterruptedException {
for (Text value : values) {
FlowBean bean = new FlowBean();
bean.set(key.getDownFlow(), key.getUpFlow());
// 1 向treeMap集合中添加数据
flowMap.put(bean, new Text(value));
// 2 限制TreeMap数据量,超过10条就删除掉流量最小的一条数据
if (flowMap.size() > 10) {
// flowMap.remove(flowMap.firstKey());
flowMap.remove(flowMap.lastKey());
}
}
}
@Override
protected void cleanup(Reducer.Context context) throws IOException, InterruptedException {
// 3 遍历集合,输出数据
Iterator it = flowMap.keySet().iterator();
while (it.hasNext()) {
FlowBean v = it.next();
context.write(new Text(flowMap.get(v)), v);
}
}
}
package com.atguigu.mr.top;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
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 TopNDriver {
public static void main(String[] args) throws Exception {
args = new String[]{"e:/output1","e:/output3"};
// 1 获取配置信息,或者job对象实例
Configuration configuration = new Configuration();
Job job = Job.getInstance(configuration);
// 6 指定本程序的jar包所在的本地路径
job.setJarByClass(TopNDriver.class);
// 2 指定本业务job要使用的mapper/Reducer业务类
job.setMapperClass(TopNMapper.class);
job.setReducerClass(TopNReducer.class);
// 3 指定mapper输出数据的kv类型
job.setMapOutputKeyClass(FlowBean.class);
job.setMapOutputValueClass(Text.class);
// 4 指定最终输出的数据的kv类型
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(FlowBean.class);
// 5 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);
}
}
MapReduce实现TopN的步骤:
(1)利用TreeMap排序, 每过来一个数据 先放入TreeMap中, 只要TreeMap的size超过n,就移除firstKey或者lastKey对应的(看是从小到大还是从大到小排序);
(2)在众多的Mapper的端,首先计算出各端Mapper的TopN,然后在将每一个Mapper端的TopN汇总到Reducer端进行计算最终的TopN,这样就可以最大化的提高运行并行处理的能力,同时极大的减少网络的Shuffle传输数据,从而极大的加快的整个处理的效率。
参考:mapreduce求topN - hdc520 - 博客园 (cnblogs.com)