我开发了一个mapReduce程序来计算和登录请求文件30分钟内的请求数以及此期间搜索次数最多的单词。
我的输入文件是：
01_11_2012 12_02_10 132.227.045.028 life
02_11_2012 02_52_10 132.227.045.028 restaurent+kitchen
03_11_2012 12_32_10 132.227.045.028 guitar+music
04_11_2012 13_52_10 132.227.045.028 book+music
05_11_2012 12_22_10 132.227.045.028 animal+life
05_11_2012 12_22_10 132.227.045.028 history
DD_MM_YYYY | HH_MM_SS | ip |搜索的单词
我的输出文件应该显示如下：
between 02h30 and 2h59 restaurent 1
between 13h30 and 13h59 book 1
between 12h00 and 12h29 life 3
between 12h30 and 12h59 guitar 1
第一行：restaurent是02h30到2h59之间最常用的词，1代表请求的数量。
我的问题是我得到同一行的redundent map执行。所以我用以下输入测试程序（我的文件中有1行）。
01_11_2012 12_02_10 132.227.045.028 life
当我每行调试eclipse行时，在下面的地图行上放置一个断点。
context.write(key, result);
我的程序在此行上传递了两次，并为唯一输入行写了两次相同的信息。
我被困在这一点上，我不知道为什么我会得到2个地图任务，因为我的输入应该只有一个分割。
该计划如下。 （对不起我的英文）
package fitec.lab.booble;
import java.io.IOException;
import java.util.Comparator;
import java.util.HashMap;
import java.util.Map;
import java.util.TreeMap;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.Text;
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;
public class BoobleByMinutes {
public static class TokenizerMapper extends Mapper<Object, Text, Text, Text> {
private final int TIME_INDEX = 1;
private final int WORDS_INDEX = 3;
@Override
public void map(Object key, Text value, Context context) throws IOException, InterruptedException {
String[] attributesTab = value.toString().split(" ");
Text reduceKey = new Text();
Text words = new Text();
String time = attributesTab[TIME_INDEX];
String[] timeSplitted = time.split("_");
String heures = timeSplitted[0];
String minutes = timeSplitted[1];
if (29 < Integer.parseInt(minutes)) {
reduceKey.set("entre " + heures + "h30 et " + heures + "h59");
} else {
reduceKey.set("entre " + heures + "h00 et " + heures + "h29");
}
words.set(attributesTab[WORDS_INDEX]);
context.write(reduceKey, words);
}
}
public static class PriceSumReducer extends Reducer<Text, Text, Text, Text> {
public void reduce(Text key, Iterable<Text> groupedWords, Context context)
throws IOException, InterruptedException {
Text result = new Text();
int requestCount = 0;
Map<String, Integer> firstWordAndRequestCount = new HashMap<String, Integer>();
for (Text words : groupedWords) {
++requestCount;
String wordsString = words.toString().replace("+", "--");
System.out.println(wordsString.toString());
String[] wordTab = wordsString.split("--");
for (String word : wordTab) {
if (firstWordAndRequestCount.containsKey(word)) {
Integer integer = firstWordAndRequestCount.get(word) + 1;
firstWordAndRequestCount.put(word, integer);
} else {
firstWordAndRequestCount.put(word, new Integer(1));
}
}
}
ValueComparator valueComparator = new ValueComparator(firstWordAndRequestCount);
TreeMap<String, Integer> sortedProductsSale = new TreeMap<String, Integer>(valueComparator);
sortedProductsSale.putAll(firstWordAndRequestCount);
result.set(sortedProductsSale.firstKey() + "__" + requestCount);
context.write(key, result);
}
class ValueComparator implements Comparator<String> {
Map<String, Integer> base;
public ValueComparator(Map<String, Integer> base) {
this.base = base;
}
public int compare(String a, String b) {
if (base.get(a) >= base.get(b)) {
return -1;
} else {
return 1;
}
}
}
}
public static void main(String[] args) throws Exception {
Job job = new org.apache.hadoop.mapreduce.Job();
job.setJarByClass(BoobleByMinutes.class);
job.setJobName("Booble mot le plus recherché et somme de requete par tranche de 30 minutes");
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
job.setJarByClass(BoobleByMinutes.class);
job.setMapperClass(TokenizerMapper.class);
// job.setCombinerClass(PriceSumReducer.class);
job.setReducerClass(PriceSumReducer.class);
job.setNumReduceTasks(1);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(Text.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
System.exit(job.waitForCompletion(true) ? 0 : 1);
}
}
@Radim 我将带有纱线的罐子放入真正的hadoop我得到的数量为split = 2
我把日志放在
之下16/07/18 02:56:39 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
16/07/18 02:56:40 INFO client.RMProxy: Connecting to ResourceManager at /0.0.0.0:8032
16/07/18 02:56:42 WARN mapreduce.JobResourceUploader: Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
16/07/18 02:56:42 INFO input.FileInputFormat: Total input paths to process : 2
16/07/18 02:56:43 INFO mapreduce.JobSubmitter: number of splits:2
16/07/18 02:56:43 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1468802929497_0002
16/07/18 02:56:44 INFO impl.YarnClientImpl: Submitted application application_1468802929497_0002
16/07/18 02:56:44 INFO mapreduce.Job: The url to track the job: http://moussa:8088/proxy/application_1468802929497_0002/
16/07/18 02:56:44 INFO mapreduce.Job: Running job: job_1468802929497_0002
16/07/18 02:56:56 INFO mapreduce.Job: Job job_1468802929497_0002 running in uber mode : false
16/07/18 02:56:56 INFO mapreduce.Job: map 0% reduce 0%
16/07/18 02:57:14 INFO mapreduce.Job: map 100% reduce 0%
16/07/18 02:57:23 INFO mapreduce.Job: map 100% reduce 100%
16/07/18 02:57:25 INFO mapreduce.Job: Job job_1468802929497_0002 completed successfully
16/07/18 02:57:25 INFO mapreduce.Job: Counters: 49
File System Counters
FILE: Number of bytes read=66
FILE: Number of bytes written=352628
FILE: Number of read operations=0
FILE: Number of large read operations=0
FILE: Number of write operations=0
HDFS: Number of bytes read=278
HDFS: Number of bytes written=31
HDFS: Number of read operations=9
HDFS: Number of large read operations=0
HDFS: Number of write operations=2
Job Counters
Launched map tasks=2
Launched reduce tasks=1
Data-local map tasks=2
Total time spent by all maps in occupied slots (ms)=29431
Total time spent by all reduces in occupied slots (ms)=6783
Total time spent by all map tasks (ms)=29431
Total time spent by all reduce tasks (ms)=6783
Total vcore-milliseconds taken by all map tasks=29431
Total vcore-milliseconds taken by all reduce tasks=6783
Total megabyte-milliseconds taken by all map tasks=30137344
Total megabyte-milliseconds taken by all reduce tasks=6945792
Map-Reduce Framework
Map input records=2
Map output records=2
Map output bytes=56
Map output materialized bytes=72
Input split bytes=194
Combine input records=0
Combine output records=0
Reduce input groups=1
Reduce shuffle bytes=72
Reduce input records=2
Reduce output records=1
Spilled Records=4
Shuffled Maps =2
Failed Shuffles=0
Merged Map outputs=2
GC time elapsed (ms)=460
CPU time spent (ms)=2240
Physical memory (bytes) snapshot=675127296
Virtual memory (bytes) snapshot=5682606080
Total committed heap usage (bytes)=529465344
Shuffle Errors
BAD_ID=0
CONNECTION=0
IO_ERROR=0
WRONG_LENGTH=0
WRONG_MAP=0
WRONG_REDUCE=0
File Input Format Counters
Bytes Read=84
File Output Format Counters
Bytes Written=31
答案 0 :(得分：1)
在您的主（作业）方法中，这些行是重复的：
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
也：job.setJarByClass(BoobleByMinutes.class);
但这一行应该导致重复输入：FileInputFormat.addInputPath(job, new Path(args[0]));
所以你的主要方法应该是：
public static void main(String[] args) throws Exception {
Job job = new org.apache.hadoop.mapreduce.Job();
job.setJarByClass(BoobleByMinutes.class);
job.setJobName("Booble mot le plus recherché et somme de requete par tranche de 30 minutes");
job.setMapperClass(TokenizerMapper.class);
// job.setCombinerClass(PriceSumReducer.class);
job.setReducerClass(PriceSumReducer.class);
job.setNumReduceTasks(1);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(Text.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
System.exit(job.waitForCompletion(true) ? 0 : 1);
}
答案 1 :(得分：0)
我从此链接获得解决方案：why is my sequence file being read twice in my hadoop mapper class?
我没有看到我得到处理的总输入路径：2在我的日志中。 正如他们在链接中所说，我只需要对该行进行评论
FileInputFormat.addInputPath(job, new Path(args[0]));
我不理解评论＆＃34;此行只是将输入添加回配置＆＃34; 请有人解释一下 任何想法表示赞赏