hadoop集群搭建及编程实践
阅读原文时间:2023年10月08日阅读:1

Hadoop集群搭建

  1. 前期准备及JDK,hadoop安装

  2. 设置主机名和添加主机映射

  3. 验证连通性

  4. SSH无密码登录

  5. 配置集群/分布式环境

  6. 分发到其他结点

  7. 格式化namenode

  8. 执行分布式实例

  9. java API与HDFS的编程

1.1JDK的下载地址,hadoop下载地址

Java Downloads | Oracle 中国

选择JDK8

清华镜像源

选择hadoop-3.3.5

注意点

查看镜像是32位还是64位

uname -m

当输出为x86_64时,说明是64位,不是的就是32位,此时需要重新下载镜像,32位不方便

1.2创建hadoop用户

在安装完linus镜像之后,需要创建一个专门的"hadoop"用户,这里的用户名为 “prettyspider"

首先按 ctrl+alt+t 打开终端窗口,输入如下命令创建新用户 :

sudo useradd -m prettyspider -s /bin/bash

-m:将prettyspider作为用户放入到用户登录目录

-s:指定用户登入后使用的shell

为用户设置登录密码

sudo passwd prettyspider

为用户添加管理员权限

sudo adduser prettyspider sudo

之后登出,登录"hadoop"用户

1.3更新apt

sudo apt-get update

同步时间

sudo apt-get install ntpdata
ntpdata -u time2.aliyun.com # 同步为阿里云NTP服务器

下载vim

sudo apt-get install vim

1.4安装SSH、配置SSH无密码登陆

sudo apt-get install openssh-server

安装完之后,登录本机

ssh localhost

在下方提示中输入yes,再根据提示输入“hadoop"用户的密码

设置免密登录之前,一定要先用密码登录一下

exit                           # 退出刚才的 ssh localhost
cd ~/.ssh/                     # 若没有该目录,请先执行一次ssh localhost
ssh-keygen -t rsa              # 会有提示,都按回车就可以
cat ./id_rsa.pub >> ./authorized_keys  # 加入授权

再使用ssh localhost登录

1.5配置远程登录

远程登录实现种类比较多,最轻便的是用vscode进行远程登录,这里使用的是MobaXterm软件

可到官网中下载MobaXterm Xserver with SSH, telnet, RDP, VNC and X11 - Download (mobatek.net)

1.6JDK安装

JDK版本为1.8.0_371

cd /usr/lib
sudo mkdir jvm #创建/usr/lib/jvm目录用来存放JDK文件
sudo tar -zxvf ~/jdk-8u371-linux-x64.tar.gz -C /usr/lib/jvm  #将

设置环境变量

cd ~
vim ~/.bashrc

在其中添加

export JAVA_HOME=/usr/lib/jvm/jdk1.8.0_371  # 对应的版本号为jdk1.8.0_对应下载版本8u后面的数字
export JRE_HOME=${JAVA_HOME}/jre
export CLASSPATH=.:${JAVA_HOME}/lib:${JRE_HOME}/lib
export PATH=${JAVA_HOME}/bin:$PATH

是配置文件生效

source ~/.bashrc

查看是否安装成功

java -version

当出现下图,表明安装成功

1.7安装hadoop

sudo tar -zxvf ~/hadoop-3.3.5.tar.gz -C /usr/local   # 解压到/usr/local中
cd /usr/local/
sudo mv ./hadoop-3.3.5/ ./hadoop            # 将文件夹名改为hadoop
sudo chown -R prettyspider ./hadoop       # 修改文件权限,prettyspider为你的”hadoop"用户名

查看hadoop是否可用

cd /usr/local/hadoop
./bin/hadoop version

出现下图,表示可用

依次配置3台主机,对应的hadoop用户名都为prettyspider

2.1修改主机名

sudo vim /etc/hostname

3台主机分别设置为 node01 node02 node03

重启后,对应的主机名便会更改,如

2.2添加主机映射

在node01结点上

sudo vim /etc/hosts

添加主机的映射,设置成下图

相应的其他结点也需要设置成上图一样

用ping指令验证连通性

ping node02 -c 3

连通成功的结果

在最开始配置的SSH是只针对当前主机而言的SSH密匙,但是不利用集群的操作,所以需要统一的配置SSH密匙

4.1在主节点上删除原有SSH,并再创建一个统一的SSH密匙

cd ~/.ssh              # 如果没有该目录,先执行一次ssh localhost
rm ./id_rsa*           # 删除之前生成的公匙(如果已经存在)
ssh-keygen -t rsa       # 执行该命令后,遇到提示信息,一直按回车就可以

将生成的密匙添加到用户的~/.ssh/authorized_keys,用于身份验证

cat ./id_rsa.pub >> ./authorized_keys

将密匙传入到对应的从结点上 传输到node02,node03

scp ~/.ssh/id_rsa.pub prettyspider@node02:/home/prettyspider/ # 此处@前后的名称为自定义的用户名和主机名 ,/home/后的为自定义的用户名

在对应的结点上实现

mkdir ~/.ssh       # 如果不存在该文件夹需先创建,若已存在,则忽略本命令
cat ~/id_rsa.pub >> ~/.ssh/authorized_keys
rm ~/id_rsa.pub    # 用完以后就可以删掉

4.3查看是否成功

ssh nod02

如下,表示成功

4.4为hadoop添加PATH

在~/.bashrc中添加

export PATH=$PATH:/usr/local/hadoop/bin:/usr/local/hadoop/sbin  # 指向对应hadoop路径下的hadoop启动文件夹的目录

5.1进入/usr/local/hadoop/etc/hadoop

/usr/local/hadoop/etc/hadoop

5.2修改workers

workers的作用:配置为DateNode的主机名,如下,删除localhost

5.3修改文件core-site.xml

指定namenode的位置和设置hadoop文件系统的基本配置

5.4修改hdfs-site.xml

配置namenode和datanode存放文件的基本路径及配置副本的数量,最小值为3

5.5修改mapred-site.xml

5.6修改yarn-site.xml

设置resourceManager运行在哪台机器上,设置NodeManager的通信方式

6.1分发其他结点

cd /usr/local
sudo rm -r ./hadoop/tmp     # 删除 Hadoop 临时文件
sudo rm -r ./hadoop/logs/*   # 删除日志文件
tar -zcf ~/hadoop.master.tar.gz ./hadoop   # 先压缩再复制
cd ~
scp ./hadoop.master.tar.gz node02:/home/prettyspider

其中

sudo rm -r ./hadoop/tmp # 删除 Hadoop 临时文件

sudo rm -r ./hadoop/logs/* # 删除日志文件

很重要,在后期配置hbase集群时有用

6.2从节点解压并设置用户组

sudo rm -r /usr/local/hadoop    # 删掉旧的(如果存在)
sudo tar -zxf ~/hadoop.master.tar.gz -C /usr/local
sudo chown -R prettyspider /usr/local/hadoop

在从结点上完成了部署hadoop,在主节点上执行名称结点的格式化

hdfs namenode -format

自此,hadoop集群搭建完成,启动集群

start-dfs.sh
start-yarn.sh
mr-jobhistory-daemon.sh start historyserver

hadoop集群的规划为

8.1创建HDFS上的用户目录

hdfs dfs -mkdir -p /user/prettyspider

hadoop用户名是什么,user后的用户就是什么

8.2创建input目录

hdfs dfs -mkdir input # input文件夹默认在用户目录下,也就是prettyspider目录下
hdfs dfs -put /usr/local/hadoop/etc/hadoop/*.xml input

8.3运行MapReduce作业

这个测试是用正则表达式获取指定前缀的任意长的字段

hadoop jar /usr/local/hadoop/share/hadoop/mapreduce/hadoop-mapreduce-examples-3.3.5.jar grep input output 'dfs[a-z.]+'

结果为

1.导入Maven依赖

<dependencies>
 &nbsp; &nbsp; <dependency>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<groupId>org.apache.hadoop</groupId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<artifactId>hadoop-common</artifactId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<version>2.7.5</version>
 &nbsp; &nbsp;</dependency>
 &nbsp; &nbsp; &nbsp; &nbsp;<dependency>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<groupId>org.apache.hadoop</groupId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<artifactId>hadoop-client</artifactId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<version>2.7.5</version>
 </dependency>
 &nbsp; &nbsp; &nbsp; &nbsp;<dependency>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<groupId>org.apache.hadoop</groupId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<artifactId>hadoop-hdfs</artifactId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<version>2.7.5</version>
 </dependency>
 <dependency>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<groupId>org.apache.hadoop</groupId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<artifactId>hadoop-mapreduce-client-core</artifactId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<version>2.7.5</version>
 </dependency>
 &nbsp; &nbsp; &nbsp; &nbsp;<dependency>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<groupId>junit</groupId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<artifactId>junit</artifactId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<version>RELEASE</version>
 &nbsp; &nbsp; &nbsp; &nbsp;</dependency>
 &nbsp; &nbsp;</dependencies>
 &nbsp; &nbsp;<build>
 &nbsp; &nbsp; &nbsp; &nbsp;<plugins>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<plugin>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<groupId>org.apache.maven.plugins</groupId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<artifactId>maven-compiler-plugin</artifactId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<version>3.1</version>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<configuration>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<source>1.8</source>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<target>1.8</target>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<encoding>UTF-8</encoding>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<!-- &nbsp; <verbal>true</verbal>-->
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;</configuration>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;</plugin>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<plugin>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<groupId>org.apache.maven.plugins</groupId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<artifactId>maven-shade-plugin</artifactId>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<version>2.4.3</version>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<executions>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<execution>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<phase>package</phase>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<goals>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<goal>shade</goal>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;</goals>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<configuration>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<minimizeJar>true</minimizeJar>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;</configuration>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;</execution>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;</executions>
 &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;</plugin>
 &nbsp; &nbsp; &nbsp; &nbsp;</plugins>
 &nbsp; &nbsp;</build>

2.上传本地文件到HDFS文件系统,将HDFS文件系统中的文件下载到本地并压缩

1.创建ConnectionJavaBean类,用于登录HDFS

package com.prettyspider.hadoop;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;

import java.io.IOException;
import java.net.URI;
import java.net.URISyntaxException;

/**
 * @author prettyspider
 * @ClassName Connection
 * @description: TODO
 * @date 2023/10/7 19:00
 * @Version V1.0
 */

public class Connection {
    // HDFS文件系统web地址
    private String hdfsUrl;
    // hadoop用户名
    private String hadoopHost;
    // 文件系统对象
    private FileSystem fs;
    public Connection() {}

    public Connection(String hdfsUrl, String hadoopHost) {
        this.hdfsUrl = hdfsUrl;
        this.hadoopHost = hadoopHost;
    }

    public Connection(String hdfsUrl, String hadoopHost, FileSystem fs) {
        this.hdfsUrl = hdfsUrl;
        this.hadoopHost = hadoopHost;
        this.fs = fs;
    }

    public String getHadoopHost() {
        return hadoopHost;
    }

    /**
     * 将web地址和hadoop用户名传入,生成文件系统对象
     * @return HDFS文件系统对象
     * @throws Exception
     */
    public FileSystem init() {
        Configuration configuration = new Configuration();
        try {
            fs = FileSystem.newInstance(new URI(hdfsUrl), configuration, hadoopHost);
        } catch (IOException e) {
            throw new RuntimeException(e);
        } catch (InterruptedException e) {
            throw new RuntimeException(e);
        } catch (URISyntaxException e) {
            throw new RuntimeException(e);
        }
        return fs;
    }

    public void fsClose() {
        try {
            fs.close();
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
    }

    /**
     * 获取
     * @return hdfsUrl
     */
    public String getHdfsUrl() {
        return hdfsUrl;
    }

    /**
     * 设置
     * @param hdfsUrl
     */
    public void setHdfsUrl(String hdfsUrl) {
        this.hdfsUrl = hdfsUrl;
    }

    /**
     * 设置
     * @param hadoopHost
     */
    public void setHadoopHost(String hadoopHost) {
        this.hadoopHost = hadoopHost;
    }

    /**
     * 获取
     * @return fs
     */
    public FileSystem getFs() {
        return fs;
    }

    /**
     * 设置
     * @param fs
     */
    public void setFs(FileSystem fs) {
        this.fs = fs;
    }

    public String toString() {
        return "Connection{hdfsUrl = " + hdfsUrl + ", hadoopHost = " + hadoopHost + ", fs = " + fs + "}";
    }
}

2.创建文件转化工具类FileTransferUtil,实现对文件夹的上传和下载

package com.prettyspider.hadoop.updateanddownload;

import org.apache.hadoop.fs.*;

import java.io.*;
import java.util.zip.ZipEntry;
import java.util.zip.ZipOutputStream;

/**
 * @author prettyspider
 * @ClassName update
 * @description: TODO
 * @date 2023/10/7 19:23
 * @Version V1.0
 */

public class FileTransferUtil {

    private FileTransferUtil() {
    }

    /**
     * 将本地指定路径下的文件上传到HDFS文件系统上
     *
     * @param localPath 本地文件路径
     * @param hdfsPath  HDFS文件系统路径
     * @param fs        HDFS文件系统对象
     */
    public static void update(String localPath, String hdfsPath, FileSystem fs) {
        /**
         * 细节:
         *      两次getName()的意义不同,第一次是获取文件夹或者文件的名称,第二次是获取文件的名称,不能共用
         */
        String name1 = new File(localPath).getName();
        hdfsPath = hdfsPath + "/" + name1;
        // 获取本地文件的文件集合
        File[] files = new File(localPath).listFiles();
        if (files != null) {
            for (File file : files) {
                // 当为文件是便上传
                if (file.isFile()) {
                    String absolutePath = file.getAbsolutePath();
                    String name = file.getName();
                    try {
                        System.out.println(hdfsPath + "/" + name);
                        fs.copyFromLocalFile(new Path("file:///" + absolutePath), new Path(hdfsPath + "/" + name));
                    } catch (IOException e) {
                        throw new RuntimeException(e);
                    }
                } else {
                    update(file.toString(), hdfsPath, fs);
                }
            }
        }
    }

    /**
     *
     * @param localPath 本地文件路径
     * @param hdfsPath HDFS文件系统路径
     * @param fs HDFS文件系统对象
     * @param username 用户名
     * @throws IOException
     */
    public static void download(String localPath, String hdfsPath, FileSystem fs,String username) throws IOException {
        RemoteIterator<LocatedFileStatus> locatedFileStatusRemoteIterator = locatedFileStatusRemoteIterator = fs.listFiles(new Path(hdfsPath), true);
        while (locatedFileStatusRemoteIterator.hasNext()) {
            LocatedFileStatus next = locatedFileStatusRemoteIterator.next();
            // 用用户名做切分点,获取从用户名开始的文件路径
            String name = next.getPath().toString().split(username)[1];
            /**
             * 细节:
             *      将获取的用户名进行切分,再组合
             */
            String[] arr = name.split("/");
            String fileName = "";
            for (int i = 0; i < arr.length - 1; i++) {
                fileName += arr[i] + "/";
            }
            // 获取HDFS文件系统的路径
            Path path = next.getPath();
            FSDataInputStream getMessage = fs.open(path);
            BufferedReader reader = new BufferedReader(new InputStreamReader(getMessage));
            /**
             * 细节:
             *      输出时需要先创建文件目录
             */
            File file = new File(localPath, fileName);
            if (!file.exists()) {
                file.mkdirs();
            }
            BufferedWriter writer = new BufferedWriter(new FileWriter(new File(file, arr[arr.length - 1])));
            String line;
            while ((line = reader.readLine()) != null) {
                writer.write(line);
                writer.newLine();
            }
            writer.close();
            reader.close();
        }

        // 压缩
        ZipOutputStream zipOutputStream = new ZipOutputStream(new FileOutputStream(new File(localPath, hdfsPath + ".zip")));
        toZIp(new File(localPath,hdfsPath), zipOutputStream, hdfsPath);
    }

    /**
     *
     * @param src 文件夹对象
     * @param zipOutputStream 压缩流
     * @param path 指定文件夹下的根目录
     * @throws IOException
     */
    private static void toZIp(File src, ZipOutputStream zipOutputStream, String path) throws IOException {
        File[] files = src.listFiles();
        if (files != null) {
            for (File file : files) {
                if (file.isFile()) {
                    ZipEntry zipEntry = new ZipEntry(path + "\\" + file.getName());
                    zipOutputStream.putNextEntry(zipEntry);
                    BufferedInputStream bufferedInputStream = new BufferedInputStream(new FileInputStream(file));
                    byte[] bytes = new byte[1024 * 1024 * 8];
                    int len;
                    while ((len = bufferedInputStream.read(bytes))!=-1) {
                        zipOutputStream.write(bytes, 0, len);
                    }
                    bufferedInputStream.close();
                } else {
                    toZIp(file, zipOutputStream, path + "\\" + file.getName());
                }
            }
            zipOutputStream.close();
        }
    }
}

测试类

FileTransferTest

import org.apache.hadoop.fs.FileSystem;
import org.junit.Test;

import java.io.IOException;

/**
 * @author prettyspider
 * @ClassName fileTransferTest
 * @description: TODO
 * @date 2023/10/7 19:47
 * @Version V1.0
 */

public class fileTransferTest {
    @Test
    public void testUpdate() throws IOException {
        Connection connection = new Connection("hdfs://node01:9000", "prettyspider");
        FileSystem fs = connection.init();
        FileTransferUtil.update("E:\\test\\wordcount","input",new ConnectionTest().testInit());
        // fileTransfer.download("E:\\test","input",fs,connection.getHadoopHost());
        connection.fsClose();
    }
}

结果

本地

HDFS文件系统Web端

3.根据HDFS文件系统查看学生是否提交作业

假设用HDFS文件系统管理学生作业,如何获取学生是否提交作业

实现:

1.根据HDFS文件系统获取指定”班级"下的所有的已经提交作业的学生

2.与班级的学生名单进行比较,获取没有提交作业的学生

创建JobSunmissionUtil工具类,实现获取没有提交做的学生

package com.prettyspider.hadoop.jobsubmission;

import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;

import java.io.*;
import java.util.ArrayList;

/**
 * @author prettyspider
 * @ClassName Search
 * @description: TODO
 * @date 2023/10/8 11:23
 * @Version V1.0
 */

public class JobSubmissionUtil {
    private JobSubmissionUtil(){}
    public static void search(FileSystem fs) throws Exception {
        File file = new File(".\\src\\main\\java\\com\\prettyspider\\hadoop\\jobsubmission\\stu.txt");
        BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(file)));
        String line;
        ArrayList<String> list = new ArrayList<>();
        ArrayList<String> nameList = new ArrayList<>();
        while ((line = reader.readLine()) != null) {
            list.add(line.split("-")[0]);
        }
        System.out.println(list);
        FileStatus[] fileStatuses = fs.listStatus(new Path("input/stu"));
        for (FileStatus fileStatus : fileStatuses) {
            String[] arr = fileStatus.getPath().toString().split("/");
            String s = arr[arr.length - 1].split("\\.")[0];
            nameList.add(s);
        }
        System.out.println(nameList);
        // 去重
        for (String name : nameList) {
            list.remove(name);
        }
        System.out.println("没有交作业的是"+list);
    }

}

测试类

JobsubmissionTest

package com.prettyspider.hadoop.updateanddownload;

import com.prettyspider.hadoop.Connection;
import com.prettyspider.hadoop.jobsubmission.JobSubmissionUtil;
import org.apache.hadoop.fs.FileSystem;
import org.junit.Test;

/**
 * @author prettyspider
 * @ClassName SearchTest
 * @description: TODO
 * @date 2023/10/8 11:30
 * @Version V1.0
 */

public class JobSubmissionTest {
    @Test
    public void testsearch() throws Exception {
        Connection connection = new Connection("hdfs://node01:9000", "prettyspider");
        FileSystem fs = connection.init();
        JobSubmissionUtil.search(fs);
        connection.fsClose();
    }
}

测试数据

4.实现HDFS文件系统指定文件夹内的文件词频统计(手搓)

MapReduce是hadoop两个核心之一,MapReduce框架由Map和Reduce组成。 Map ()负责把一个大的block块进行切片并计算。 Reduce () 负责把Map ()切片的数据进行汇总、计算。

那么可以通过简化,实现切片和数据统计

实现步骤:

1.将HDFS文件系统指定文件夹下的文件合并到一个文件中

2.对文件进行切分

3.将切分之后的数据利用Map集合实现统计

创建WordCountUtil工具类

package com.prettyspider.hadoop.wordcount;

import org.apache.hadoop.fs.*;

import java.io.*;
import java.util.*;

/**
 * @author prettyspider
 * @ClassName wordcount
 * @description: TODO
 * @date 2023/10/8 12:46
 * @Version V1.0
 */

public class WordCountUtil {
    private WordCountUtil() {}

    /**
     * 将指定文件夹下的文件合并到一个文件中,再对文件进行词频统计
     * @param fs HDFS文件系统对象
     * @param hdfsPath 要统计词频的文件夹地址
     * @param mergePath 合并后的文件地址
     * @throws IOException
     */
    public static void wordcount(FileSystem fs,String hdfsPath,String mergePath) throws IOException {
        merge(fs, hdfsPath, mergePath);
        wordcount(fs, mergePath);
    }

    /**
     * 利用Map对数据进行统计
     * @param fs HDFS文件系统
     * @param mergePath 合并的文件地址
     * @throws IOException
     */
    private static void wordcount(FileSystem fs, String mergePath) throws IOException {
        FSDataInputStream open = fs.open(new Path(mergePath));
        // 用集合获取数据
        ArrayList<String> list = new ArrayList<>();
        BufferedReader reader = new BufferedReader(new InputStreamReader(open));
        String line;
        while ((line = reader.readLine()) != null) {
            list.add(line);
        }
        StringBuilder stringBuilder = new StringBuilder();
        for (String s : list) {
            stringBuilder.append(s);
        }
        String[] arr = stringBuilder.toString().split("\\W+");
        // 词频统计
        wordstatistic(arr);
    }

    /**
     *
     * @param arr 被拆分后的词的数组
     */
    private static void wordstatistic(String[] arr) {
        HashMap<String, Integer> map = new HashMap<>();
        for (int i = 0; i < arr.length; i++) {
            String s = arr[i];
            // map中不存在数据
            if (!map.containsKey(s)) {
                map.put(s, 1);
            } else {
                int count = map.get(s) + 1;
                map.put(s,count);
            }
        }
        // 输出结果
        Set<Map.Entry<String, Integer>> entries = map.entrySet();
        for (Map.Entry<String, Integer> entry : entries) {
            String key = entry.getKey();
            Integer value = entry.getValue();
            System.out.println("key="+key+",value="+value);
        }
    }

    /**
     *
     * @param fs HDFS文件系统对象
     * @param hdfsPath 要统计的文件夹地址
     * @param mergePath 合并后文件地址
     * @throws IOException
     */
    private static void merge(FileSystem fs, String hdfsPath, String mergePath) throws IOException {
        FSDataOutputStream fsDataOutputStream = fs.create(new Path(mergePath));
        FileStatus[] fileStatuses = fs.listStatus(new Path(hdfsPath));
        BufferedWriter writer = new BufferedWriter(new OutputStreamWriter(fsDataOutputStream));
        for (FileStatus fileStatus : fileStatuses) {
            FSDataInputStream open = fs.open(new Path(fileStatus.getPath().toUri()));
            BufferedReader reader = new BufferedReader(new InputStreamReader(open));
            String line;
            while ((line = reader.readLine()) != null) {
                writer.write(line);
                writer.newLine();
            }
            reader.close();
            open.close();
        }
        writer.close();
        fsDataOutputStream.close();
    }

}

测试类

WordCountTest

package com.prettyspider.hadoop.updateanddownload;

import com.prettyspider.hadoop.Connection;
import com.prettyspider.hadoop.wordcount.WordCountUtil;
import org.apache.hadoop.fs.FileSystem;
import org.junit.Test;

import java.io.IOException;

/**
 * @author prettyspider
 * @ClassName WordCountTest
 * @description: TODO
 * @date 2023/10/8 13:15
 * @Version V1.0
 */

public class WordCountTest {
    @Test
    public void testwordcount() throws IOException {
        Connection connection = new Connection("hdfs://node01:9000", "prettyspider");
        FileSystem fs = connection.init();
        WordCountUtil.wordcount(fs,"input/wordcount","output/merge.txt");
        connection.fsClose();
    }
}

结果

手机扫一扫

移动阅读更方便

阿里云服务器
腾讯云服务器
七牛云服务器

你可能感兴趣的文章