package com.cakin.hadoop.mr;
import java.io.DataInput;
import java.io.DataOutput;
import java.io.IOException;
import org.apache.hadoop.io.WritableComparable;
public class UserWritable implements WritableComparable<UserWritable> {
private Integer id;
private Integer income;
private Integer expenses;
private Integer sum;
public void write(DataOutput out) throws IOException {
// TODO Auto-generated method stub
out.writeInt(id);
out.writeInt(income);
out.writeInt(expenses);
out.writeInt(sum);
}
public void readFields(DataInput in) throws IOException {
// TODO Auto-generated method stub
this.id=in.readInt();
this.income=in.readInt();
this.expenses=in.readInt();
this.sum=in.readInt();
}
public Integer getId() {
return id;
}
public UserWritable setId(Integer id) {
this.id = id;
return this;
}
public Integer getIncome() {
return income;
}
public UserWritable setIncome(Integer income) {
this.income = income;
return this;
}
public Integer getExpenses() {
return expenses;
}
public UserWritable setExpenses(Integer expenses) {
this.expenses = expenses;
return this;
}
public Integer getSum() {
return sum;
}
public UserWritable setSum(Integer sum) {
this.sum = sum;
return this;
}
public int compareTo(UserWritable o) {
// TODO Auto-generated method stub
return this.id>o.getId()?1:-1;
}
@Override
public String toString() {
return id + "\t"+income+"\t"+expenses+"\t"+sum;
}
}
package com.cakin.hadoop.mr;
import java.io.IOException;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.NullWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapreduce.Job;
import org.apache.hadoop.mapreduce.Mapper;
import org.apache.hadoop.mapreduce.lib.input.FileInputFormat;
import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
import org.apache.hadoop.mapreduce.Reducer;
/*
* 测试数据
* 用户id 收入 支出
* 1 1000 0
* 2 500 300
* 1 2000 1000
* 2 500 200
*
* 需求:
* 用户id 总收入 总支出 总的余额
* 1 3000 1000 2000
* 2 1000 500 500
* */
public class CountMapReduce {
public static class CountMapper extends Mapper<LongWritable,Text,IntWritable,UserWritable>
{
private UserWritable userWritable =new UserWritable();
private IntWritable id =new IntWritable();
@Override
protected void map(LongWritable key,Text value,
Mapper<LongWritable,Text,IntWritable,UserWritable>.Context context) throws IOException, InterruptedException{
String line = value.toString();
String[] words = line.split("\t");
if(words.length ==3)
{
userWritable.setId(Integer.parseInt(words[0]))
.setIncome(Integer.parseInt(words[1]))
.setExpenses(Integer.parseInt(words[2]))
.setSum(Integer.parseInt(words[1])-Integer.parseInt(words[2]));
id.set(Integer.parseInt(words[0]));
}
context.write(id, userWritable);
}
}
public static class CountReducer extends Reducer<IntWritable,UserWritable,UserWritable,NullWritable>
{
/*
* 输入数据
* <1,{[1,1000,0,1000],[1,2000,1000,1000]}>
* <2,[2,500,300,200],[2,500,200,300]>
*
* */
private UserWritable userWritable = new UserWritable();
private NullWritable n = NullWritable.get();
protected void reduce(IntWritable key,Iterable<UserWritable> values,
Reducer<IntWritable,UserWritable,UserWritable,NullWritable>.Context context) throws IOException, InterruptedException{
Integer income=0;
Integer expenses = 0;
Integer sum =0;
for(UserWritable u:values)
{
income += u.getIncome();
expenses+=u.getExpenses();
}
sum = income - expenses;
userWritable.setId(key.get())
.setIncome(income)
.setExpenses(expenses)
.setSum(sum);
context.write(userWritable, n);
}
}
public static void main(String[] args) throws IllegalArgumentException, IOException, ClassNotFoundException, InterruptedException {
Configuration conf=new Configuration();
/*
* 集群中节点都有配置文件
conf.set("mapreduce.framework.name.", "yarn");
conf.set("yarn.resourcemanager.hostname", "mini1");
*/
Job job=Job.getInstance(conf,"countMR");
//jar包在哪里,现在在客户端,传递参数
//任意运行,类加载器知道这个类的路径,就可以知道jar包所在的本地路径
job.setJarByClass(CountMapReduce.class);
//指定本业务job要使用的mapper/Reducer业务类
job.setMapperClass(CountMapper.class);
job.setReducerClass(CountReducer.class);
//指定mapper输出数据的kv类型
job.setMapOutputKeyClass(IntWritable.class);
job.setMapOutputValueClass(UserWritable.class);
//指定最终输出的数据kv类型
job.setOutputKeyClass(UserWritable.class);
job.setOutputKeyClass(NullWritable.class);
//指定job的输入原始文件所在目录
FileInputFormat.setInputPaths(job, new Path(args[0]));
//指定job的输出结果所在目录
FileOutputFormat.setOutputPath(job, new Path(args[1]));
//将job中配置的相关参数及job所用的java类在的jar包,提交给yarn去运行
//提交之后,此时客户端代码就执行完毕,退出
//job.submit();
//等集群返回结果在退出
boolean res=job.waitForCompletion(true);
System.exit(res?0:1);
//类似于shell中的$?
}
}
[root@centos hadoop-2.7.4]# bin/hdfs dfs -cat /input/data
1 1000 0
2 500 300
1 2000 1000
2 500 200
[root@centos hadoop-2.7.4]# bin/yarn jar /root/jar/mapreduce.jar /input/data /output3
17/12/20 21:24:45 INFO client.RMProxy: Connecting to ResourceManager at /0.0.0.0:8032
17/12/20 21:24:46 WARN mapreduce.JobResourceUploader: Hadoop command-line option parsing not performed. Implement the Tool interface and execute your application with ToolRunner to remedy this.
17/12/20 21:24:47 INFO input.FileInputFormat: Total input paths to process : 1
17/12/20 21:24:47 INFO mapreduce.JobSubmitter: number of splits:1
17/12/20 21:24:47 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1513775596077_0001
17/12/20 21:24:49 INFO impl.YarnClientImpl: Submitted application application_1513775596077_0001
17/12/20 21:24:49 INFO mapreduce.Job: The url to track the job: http://centos:8088/proxy/application_1513775596077_0001/
17/12/20 21:24:49 INFO mapreduce.Job: Running job: job_1513775596077_0001
17/12/20 21:25:13 INFO mapreduce.Job: Job job_1513775596077_0001 running in uber mode : false
17/12/20 21:25:13 INFO mapreduce.Job: map 0% reduce 0%
17/12/20 21:25:38 INFO mapreduce.Job: map 100% reduce 0%
17/12/20 21:25:54 INFO mapreduce.Job: map 100% reduce 100%
17/12/20 21:25:56 INFO mapreduce.Job: Job job_1513775596077_0001 completed successfully
17/12/20 21:25:57 INFO mapreduce.Job: Counters: 49
File System Counters
FILE: Number of bytes read=94
FILE: Number of bytes written=241391
FILE: Number of read operations=0
FILE: Number of large read operations=0
FILE: Number of write operations=0
HDFS: Number of bytes read=135
HDFS: Number of bytes written=32
HDFS: Number of read operations=6
HDFS: Number of large read operations=0
HDFS: Number of write operations=2
Job Counters
Launched map tasks=1
Launched reduce tasks=1
Data-local map tasks=1
Total time spent by all maps in occupied slots (ms)=23672
Total time spent by all reduces in occupied slots (ms)=11815
Total time spent by all map tasks (ms)=23672
Total time spent by all reduce tasks (ms)=11815
Total vcore-milliseconds taken by all map tasks=23672
Total vcore-milliseconds taken by all reduce tasks=11815
Total megabyte-milliseconds taken by all map tasks=24240128
Total megabyte-milliseconds taken by all reduce tasks=12098560
Map-Reduce Framework
Map input records=4
Map output records=4
Map output bytes=80
Map output materialized bytes=94
Input split bytes=94
Combine input records=0
Combine output records=0
Reduce input groups=2
Reduce shuffle bytes=94
Reduce input records=4
Reduce output records=2
Spilled Records=8
Shuffled Maps =1
Failed Shuffles=0
Merged Map outputs=1
GC time elapsed (ms)=157
CPU time spent (ms)=1090
Physical memory (bytes) snapshot=275660800
Virtual memory (bytes) snapshot=4160692224
Total committed heap usage (bytes)=139264000
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=41
File Output Format Counters
Bytes Written=32
[root@centos hadoop-2.7.4]# bin/hdfs dfs -cat /output3/part-r-00000
1 3000 1000 2000
2 1000 500 500
文章浏览阅读3.6k次。需求背景EasyDarwin开发团队维护EasyDarwin开源流媒体服务器也已经很多年了,之前也陆陆续续尝试过很多种服务端录像的方案,有:在EasyDarwin中直接解析收到的RTP包,重新组包录像;也有:在EasyDarwin中新增一个RecordModule,再以RTSPClient的方式请求127.0.0.1自己的直播流录像,但这些始终都没有成气候;我们的想法是能够让整套EasyDarwin_开源录播系统
文章浏览阅读1.1w次。今天碰到一个执行语句等了半天没有执行:delete table XXX where ......,但是在select 的时候没问题。后来发现是在执行select * from XXX for update 的时候没有commit,oracle将该记录锁住了。可以通过以下办法解决: 先查询锁定记录 Sql代码 SELECT s.sid, s.seri_oracle delete update 锁表问题
文章浏览阅读3.4k次。报错信息error:Undefined symbol: typeinfo for sdk::IConfigUndefined symbol: vtable for sdk::IConfig具体信息:Undefined symbols for architecture x86_64: "typeinfo for sdk::IConfig", referenced from: typeinfo for sdk::ConfigImpl in sdk.a(config_impl.o) _xcode undefined symbols:
文章浏览阅读249次。背景《承接上文,项目05(Mysql升级06Mysql5.6.51升级到Mysql5.7.32)》,写在前面需要(考虑)检查和测试的层面很多,不限于以下内容。参考文档https://dev.mysql.com/doc/refman/8.0/en/upgrade-prerequisites.htmllink推荐阅读以上链接,因为对应以下问题,有详细的建议。官方文档:不得存在以下问题:0.不得有使用过时数据类型或功能的表。不支持就地升级到MySQL 8.0,如果表包含在预5.6.4格_mysql8.0.26 升级32
文章浏览阅读3.7k次。一.安装基本环境工具:1.安装git工具sudo apt install wget g++ git2.检查并安装java等环境工具2.1、执行下面安装命令#!/bin/bashsudoapt-get-yinstall--upgraderarunrarsudoapt-get-yinstall--upgradepython-pippython3-pip#aliyunsudoapt-get-yinstall--upgradeopenjdk..._高通8155 qnx 源码
文章浏览阅读461次。firebase 与谷歌 大多数开发人员都听说过Google的Firebase产品。 这就是Google所说的“ 移动平台,可帮助您快速开发高质量的应用程序并发展业务。 ”。 它基本上是大多数开发人员在构建应用程序时所需的一组工具。 在本文中,我将介绍这些工具,并指出您选择使用Firebase时需要了解的所有内容。 在开始之前,我需要说的是,我不会详细介绍Firebase提供的所有工具。 我..._firsebase 与 google
文章浏览阅读1.2k次。在容器化应用中,每个环境都要独立的打一个镜像再给镜像一个特有的tag,这很麻烦,这就要用到k8s原生的配置中心configMap就是用解决这个问题的。使用configMap部署应用。这里使用nginx来做示例,简单粗暴。直接用vim常见nginx的配置文件,用命令导入进去kubectl create cm nginx.conf --from-file=/home/nginx.conf然后查看kub..._pod mount目录会自动创建吗
文章浏览阅读169次。随着互联网技术的发发展,计算机技术广泛应用在人们的生活中,逐渐成为日常工作、生活不可或缺的工具,高校各种管理系统层出不穷。高校作为学习知识和技术的高等学府,信息技术更加的成熟,为新生报到管理开发必要的系统,能够有效的提升管理效率。一直以来,新生报到一直没有进行系统化的管理,学生无法准确查询学院信息,高校也无法记录新生报名情况,由此提出开发基于微服务的分布式新生报到系统,管理报名信息,学生可以在线查询报名状态,节省时间,提高效率。_关于spring cloud的参考文献有啥
文章浏览阅读3.2k次。Public MustInherit Class Contact '只能作基类且不能实例化 Private mID As Guid = Guid.NewGuid Private mName As String Public Property ID() As Guid Get Return mID End Get_vb.net 继承多个接口
文章浏览阅读1.7k次。1.美图# 2.概述因为要上传我的所有仓库的包,希望nexus中已有的包,我不覆盖,没有的添加。所以想批量上传jar。3.方案1-脚本批量上传PS:nexus3.x版本只能通过脚本上传3.1 批量放入jar在mac目录下,新建一个文件夹repo,批量放入我们需要的本地库文件夹,并对文件夹授权(base) lcc@lcc nexus-3.22.0-02$ mkdir repo2..._nexus3 批量上传jar包 java代码
文章浏览阅读6.6k次,点赞6次,收藏30次。本文转自http://blog.csdn.net/charleslei/article/details/486519531、什么是场在介绍Deinterlacer去隔行处理的方法之前,我们有必要提一下关于交错场和去隔行处理的基本知识。那么什么是场呢,场存在于隔行扫描记录的视频中,隔行扫描视频的每帧画面均包含两个场,每一个场又分别含有该帧画面的奇数行扫描线或偶数行扫描线信息,_mipi去隔行
文章浏览阅读1.7k次。DATA L_ENDDA TYPE SY-DATUM. IF P_DATE IS INITIAL. CONCATENATE SY-DATUM(4) '1231' INTO L_ENDDA. ELSE. CONCATENATE P_DATE(4) '1231' INTO L_ENDDA. ENDIF. DATA: LV_RESET(1) TY_abap 自定义 search help