tencent cloud

Description

Hadoop-COS implements a standard Hadoop file systems on the COS platform. It helps integrate COS with big data computing frameworks such as Hadoop, Spark, and Tez, so that they can read and write COS data as they do with HDFS file systems.

Since Hadoop-COS uses COSN (a Tencent Cloud big data tool) as its URI scheme, it can also be referred to as a COSN-based file system.

Operating Environments

Operating system

Windows, Linux, or macOS

Software dependency

Hadoop-2.6.0 or later

Note：

1. Hadoop-COS has been integrated in Apache Hadoop-3.3.0. For more information, click here.
2. If your version is earlier than Apache Hadoop-3.3.0, or the CDH has integrated the Hadoop-COS JAR package, you need to restart NodeManager to load the JAR package.
3. To build a JAR package of a specified Hadoop version, modify hadoop.version in the pom file.

Download and Installation

Obtaining Hadoop-COS JAR package and dependencies

Download link: Hadoop-COS release

Installing Hadoop-COS plugin

1. Copy hadoop-cos-{hadoop.version}-{version}.jar and cos_api-bundle-{version}.jar to $HADOOP_HOME/share/hadoop/tools/lib.

   Note：

   Select the JAR package that corresponds to your Hadoop version. If you cannot find the desired JAR package in the release, manually build and generate one by modifying hadoop.version in the POM file.

2. Modify the hadoop-env.sh file under the $HADOOP_HOME/etc/hadoop directory by adding the COSN JAR package to your Hadoop environment variables as follows:

   for f in $HADOOP_HOME/share/hadoop/tools/lib/*.jar; do
   if [ "$HADOOP_CLASSPATH" ]; then
   export HADOOP_CLASSPATH=$HADOOP_CLASSPATH:$f
   else
   export HADOOP_CLASSPATH=$f
   fi
   done
   

Configuration Method

COSN configuration

Attribute Key	Description	Default Value	Required
fs.cosn.userinfo. secretId/secretKey	Specifies the API key for your account. Log in to the CAM console to view the key.	None	Yes
fs.cosn. credentials.provider	Specifies how to obtain your SecretId and SecretKey. There are eight methods to choose from: org.apache.hadoop.fs.auth.SessionCredentialProvider: Obtains them from the request URI. Format: cosn://{secretId}:{secretKey}@examplebucket-1250000000/. org.apache.hadoop.fs.auth.SimpleCredentialProvider: Obtains them by reading fs.cosn.userinfo.secretId and fs.cosn.userinfo.secretKey form the core-site.xml configuration file. org.apache.hadoop.fs.auth.EnvironmentVariableCredentialProvider: Obtains them from the system environment variables COS_SECRET_ID and COS_SECRET_KEY. org.apache.hadoop.fs.auth.SessionTokenCredentialProvider: Uses a temporary key for access. org.apache.hadoop.fs.auth.CVMInstanceCredentialsProvider: Obtains a temporary key for COS access by using a CVM-bound role. org.apache.hadoop.fs.auth.CPMInstanceCredentialsProvider: Obtains a temporary key for COS access by using a CPM-bound role. org.apache.hadoop.fs.auth.EMRInstanceCredentialsProvider: Obtains a temporary key for COS access by using an EMR-bound role. org.apache.hadoop.fs.auth.RangerCredentialsProvider: Obtains the key by using Ranger.	If this parameter is not specified, the default order is as follows: org.apache.hadoop.fs.auth.SessionCredentialProvider org.apache.hadoop.fs.auth.SimpleCredentialProvider org.apache.hadoop.fs.auth.EnvironmentVariableCredentialProvider org.apache.hadoop.fs.auth.SessionTokenCredentialProvider org.apache.hadoop.fs.auth.CVMInstanceCredentialsProvider org.apache.hadoop.fs.auth.CPMInstanceCredentialsProvider org.apache.hadoop.fs.auth.EMRInstanceCredentialsProvider.	No
fs.cosn.useHttps	Indicates whether to use HTTPS as the transfer protocol for the COS backend.	false	No
fs.cosn.impl	COSN implementation class for FileSystem; fixed as org.apache.hadoop.fs.CosFileSystem.	None	Yes
fs.AbstractFileSystem. cosn.impl	COSN implementation class for AbstractFileSystem; fixed as org.apache.hadoop.fs.CosN.	None	Yes
fs.cosn.bucket.region	Specifies the region of the bucket to access. For enumerated values, see the region abbreviations outlined in Regions and Access Endpoints. Example: ap-beijing and ap-guangzhou. Compatibility is provided for the old parameter fs.cosn.userinfo.region.	None	Yes
fs.cosn.bucket. endpoint_suffix	Specifies the COS endpoint to connect (optional). Public cloud COS users only need to provide the correct region in the parameter above. Compatibility is provided for the old parameter fs.cosn.userinfo.endpoint_suffix.	None	No
fs.cosn.tmp.dir	Specifies an existing local directory where temporary files generated at runtime are stored.	/tmp/hadoop_cos	No
fs.cosn.upload. part.size	The size of each part for multipart upload using the COSN file system. A COS multipart upload supports a maximum of 10,000 parts to be uploaded for a single object. You need to estimate the desired part size as needed. For example, if the part size is set to 8388608 (8 MB), you can upload an object of up to 78 GB. The size of a part can be up to 2 GB, that is, the size of a single object can be up to 19 TB.	8388608 (8 MB)	No
fs.cosn. upload.buffer	The type of buffer used when uploading files with COSN. Currently, there are three types of buffers: non_direct_memory, direct_memory, and mapped_disk. The non-direct memory buffer uses JVM on-heap memory, the direct_memory buffer uses off-heap memory, and the mapped_disk buffer works based on memory file mapping.	mapped_disk	No
fs.cosn. upload.buffer.size	The size of buffer used during the upload with COSN. A value of -1 means there is no limit. You can specify this value only if you set the buffer type to mapped_disk. If you specify a value greater than 0, it cannot be smaller than the block size. Compatibility is provided for the old parameter fs.cosn.buffer.size.	-1	No
fs.cosn.block.size	The size of a block in the COSN file system.	134217728 (128 MB)	No
fs.cosn. upload_thread_pool	Specifies the number of concurrent threads when files are uploaded to COS via streams.	10	No
fs.cosn. copy_thread_pool	Specifies the number of threads used to copy and delete concurrent files during directory replication.	3	No
fs.cosn. read.ahead.block.size	Size of the read-ahead block	1048576 (1 MB)	No
fs.cosn. read.ahead.queue.size	Length of the read-ahead queue	8	No
fs.cosn.maxRetries	The maximum number of retries if an error occurs when accessing COS	200	No
fs.cosn.retry. interval.seconds	Time interval between retries in seconds	3	No
fs.cosn. server-side-encryption.algorithm	Specifies the COS server-side encryption algorithm. Valid values: SSE-C and SSE-COS. If this parameter is left empty (default), no encryption algorithm is used.	None	No
fs.cosn. server-side-encryption.key	Specifies the required SSE-C key if the SSE-C server encryption algorithm is used. This parameter is a Base64-encoded AES-256 key. If this parameter is left empty (default), no encryption key is used.	None	No
fs.cosn. crc64.checksum.enabled	Indicates whether to enable CRC-64 checksum. It is disabled by default, meaning that you can’t run the hadoop fs -checksum command to obtain the CRC-64 checksum of a file.	false	No
fs.cosn. crc32c.checksum.enabled	Indicates whether to enable CRC32C checksum. It is disabled by default, meaning that you cannot run the hadoop fs -checksum command to obtain the CRC32C checksum of a file. CRC-64 and CRC32C cannot be both enabled.	false	No
fs.cosn.traffic.limit	Specifies the limit on the upload bandwidth, in bit/s. Value range: 819200-838860800. Defaults to -1, meaning no limit.	None	No

Hadoop configuration

Modify $HADOOP_HOME/etc/hadoop/core-site.xml by adding information on COS users and implementation classes, as shown below:

<configuration>
  <property>
      <name>fs.cosn.credentials.provider</name>
      <value>org.apache.hadoop.fs.auth.SimpleCredentialProvider</value>
      <description>
           This option allows the user to specify how to get the credentials.
          Comma-separated class names of credential provider classes which implement
          com.qcloud.cos.auth.COSCredentialsProvider:
           1.org.apache.hadoop.fs.auth.SessionCredentialProvider: Obtain the secret id and secret key from the URI: cosn://secretId:secretKey@examplebucket-1250000000/;
          2.org.apache.hadoop.fs.auth.SimpleCredentialProvider: Obtain the secret id and secret key
          from fs.cosn.userinfo.secretId and fs.cosn.userinfo.secretKey in core-site.xml;
          3.org.apache.hadoop.fs.auth.EnvironmentVariableCredentialProvider: Obtain the secret id and secret key
          from system environment variables named COS_SECRET_ID and COS_SECRET_KEY.
           If unspecified, the default order of credential providers is:
          1. org.apache.hadoop.fs.auth.SessionCredentialProvider
          2. org.apache.hadoop.fs.auth.SimpleCredentialProvider
          3. org.apache.hadoop.fs.auth.EnvironmentVariableCredentialProvider
          4. org.apache.hadoop.fs.auth.SessionTokenCredentialProvider
          5. org.apache.hadoop.fs.auth.CVMInstanceCredentialsProvider
          6. org.apache.hadoop.fs.auth.CPMInstanceCredentialsProvider
          7. org.apache.hadoop.fs.auth.EMRInstanceCredentialsProvider
      </description>
  </property>

  <property>
      <name>fs.cosn.userinfo.secretId</name>
      <value>xxxxxxxxxxxxxxxxxxxxxxxxx</value>
      <description>Tencent Cloud Secret Id</description>
  </property>
    
  <property>
      <name>fs.cosn.userinfo.secretKey</name>
      <value>xxxxxxxxxxxxxxxxxxxxxxxx</value>
      <description>Tencent Cloud Secret Key</description>
  </property>
    
  <property>
      <name>fs.cosn.bucket.region</name>
      <value>ap-xxx</value>
      <description>The region where the bucket is located.</description>
  </property>
    
  <property>
      <name>fs.cosn.bucket.endpoint_suffix</name>
      <value>cos.ap-xxx.myqcloud.com</value>
      <description>
        COS endpoint to connect to. 
        For public cloud users, it is recommended not to set this option, and only the correct area field is required.
      </description>
  </property>
    
  <property>
      <name>fs.cosn.impl</name>
      <value>org.apache.hadoop.fs.CosFileSystem</value>
      <description>The implementation class of the CosN Filesystem.</description>
  </property>
    
  <property>
      <name>fs.AbstractFileSystem.cosn.impl</name>
      <value>org.apache.hadoop.fs.CosN</value>
      <description>The implementation class of the CosN AbstractFileSystem.</description>
  </property>
   <property>
      <name>fs.cosn.tmp.dir</name>
      <value>/tmp/hadoop_cos</value>
      <description>Temporary files will be placed here.</description>
  </property>
   <property>
      <name>fs.cosn.upload.buffer</name>
      <value>mapped_disk</value>
      <description>The type of upload buffer. Available values: non_direct_memory, direct_memory, mapped_disk</description>
  </property>
   <property>
      <name>fs.cosn.upload.buffer.size</name>
      <value>134217728</value>
      <description>The total size of the upload buffer pool. -1 means unlimited.</description>
  </property>
    <property>
      <name>fs.cosn.upload.part.size</name>
      <value>8388608</value>
      <description>Block size to use cosn filesysten, which is the part size for MultipartUpload.
      Considering the COS supports up to 10000 blocks, user should estimate the maximum size of a single file.
      For example, 8MB part size can allow  writing a 78GB single file.</description>
  </property>
    
  <property>
      <name>fs.cosn.maxRetries</name>
    <value>3</value>
    <description>
      The maximum number of retries for reading or writing files to
      COS, before we signal failure to the application.
    </description>
  </property>
    
  <property>
      <name>fs.cosn.retry.interval.seconds</name>
    <value>3</value>
    <description>The number of seconds to sleep between each COS retry.</description>
  </property>
 
  <property>
      <name>fs.cosn.server-side-encryption.algorithm</name>
      <value></value>
      <description>The server side encryption algorithm.</description>
  </property>    
     <property>
      <name>fs.cosn.server-side-encryption.key</name>
      <value></value>
      <description>The SSE-C server side encryption key.</description>
  </property> 
    
</configuration>

We recommend you not configure fs.defaultFS in the production environment. To use it for certain test cases (for example, hive-testbench), add the following configurations:

<property>
        <name>fs.defaultFS</name>
        <value>cosn://examplebucket-1250000000</value>
      <description>
           This option is not advice to config, this only used for some special test cases.
      </description>
</property>

Server-side encryption

Hadoop-COS supports server-side encryption using either of the following two methods: COS-managed keys (SSE-COS) and customer-provided keys (SSE-C). You can enable this feature, which is disabled by default, by configuring as shown below:

SSE-COS encryption

SSE-COS encryption refers to server-side encryption with COS-managed keys. That is, Tencent Cloud COS manages the master key and its data. When using Hadoop-COS, you can add the following configuration in the $ HADOOP_HOME/etc/hadoop/core-site.xml file to implement SSE-COS encryption.

<property>
      <name>fs.cosn.server-side-encryption.algorithm</name>
      <value>SSE-COS</value>
      <description>The server side encryption algorithm.</description>
</property>

SSE-C encryption

SSE-C encryption refers to server-side encryption with customer-provided keys. That is, the encryption key is provided by the user. When you upload an object, COS will use the encryption key that you provide to apply AES-256 encryption to the data. When using Hadoop-COS, you can add the following configuration in the $ HADOOP_HOME/etc/hadoop/core-site.xml file to implement SSE-C encryption.

<property>
      <name>fs.cosn.server-side-encryption.algorithm</name>
      <value>SSE-C</value>
      <description>The server side encryption algorithm.</description>
</property>        
<property>
    <name>fs.cosn.server-side-encryption.key</name>
      <value>MDEyMzQ1Njc4OUFCQ0RFRjAxMjM0NTY3ODlBQkNERUY=</value> #Users need to configure the SSE-C key in the format of a Base64-encoded AES-256 key.
      <description>The SSE-C server side encryption key.</description>
</property> 

Note：

• The SSE-C encryption feature of Hadoop-COS relies on the SSE-C server-side encryption of COS, which means Hadoop-COS does not store any customer-provided encryption keys, just like COS. Instead, COS stores HMAC values with random data added to the encryption keys to verify access requests. COS cannot use the HMAC values to derive the value of an encryption key or decrypt object content. Therefore, if you lose your encryption key, you will not be able to access the object again.
• If you configure an SSE-C server-side encryption algorithm on Hadoop-COS, you must also configure an SSE-C key using the fs.cosn.server-side-encryption.key parameter in the format of a Base64-encoded AES-256 key.

How to Use

Samples

Run a command in the format: hadoop fs -ls -R cosn://<bucketname-appid>/<path> or hadoop fs -ls -R /<path> (you need to set fs.defaultFS to cosn://BucketName-APPID). The following sample uses a bucket named examplebucket-1250000000, to which you can append a specific path.

hadoop fs -ls -R cosn://examplebucket-1250000000/
-rw-rw-rw-   1 root root       1087 2018-06-11 07:49 cosn://examplebucket-1250000000/LICENSE
drwxrwxrwx   - root root          0 1970-01-01 00:00 cosn://examplebucket-1250000000/hdfs
drwxrwxrwx   - root root          0 1970-01-01 00:00 cosn://examplebucket-1250000000/hdfs/2018
-rw-rw-rw-   1 root root       1087 2018-06-12 03:26 cosn://examplebucket-1250000000/hdfs/2018/LICENSE
-rw-rw-rw-   1 root root       2386 2018-06-12 03:26 cosn://examplebucket-1250000000/hdfs/2018/ReadMe
drwxrwxrwx   - root root          0 1970-01-01 00:00 cosn://examplebucket-1250000000/hdfs/test
-rw-rw-rw-   1 root root       1087 2018-06-11 07:32 cosn://examplebucket-1250000000/hdfs/test/LICENSE
-rw-rw-rw-   1 root root       2386 2018-06-11 07:29 cosn://examplebucket-1250000000/hdfs/test/ReadMe

Run the wordcount provided in MapReduce and execute the following command.

Note：

This sample uses hadoop-mapreduce-examples-2.7.2.jar. To use a different version of the JAR file, modify the version number.

bin/hadoop jar share/hadoop/mapreduce/hadoop-mapreduce-examples-2.7.2.jar wordcount cosn://example/mr/input cosn://example/mr/output3

If the command is successful, the following information will be returned:

File System Counters
COSN: Number of bytes read=72
COSN: Number of bytes written=40
COSN: Number of read operations=0
COSN: Number of large read operations=0
COSN: Number of write operations=0
FILE: Number of bytes read=547350
FILE: Number of bytes written=1155616
FILE: Number of read operations=0
FILE: Number of large read operations=0
FILE: Number of write operations=0
HDFS: Number of bytes read=0
HDFS: Number of bytes written=0
HDFS: Number of read operations=0
HDFS: Number of large read operations=0
HDFS: Number of write operations=0
Map-Reduce Framework
Map input records=5
Map output records=7
Map output bytes=59
Map output materialized bytes=70
Input split bytes=99
Combine input records=7
Combine output records=6
Reduce input groups=6
Reduce shuffle bytes=70
Reduce input records=6
Reduce output records=6
Spilled Records=12
Shuffled Maps =1
Failed Shuffles=0
Merged Map outputs=1
GC time elapsed (ms)=0
Total committed heap usage (bytes)=653262848
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=36
File Output Format Counters
Bytes Written=40

Accessing COSN via Java code

package com.qcloud.chdfs.demo;
import org.apache.commons.io.IOUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileChecksum;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import java.io.IOException;
import java.net.URI;
import java.nio.ByteBuffer;
public class Demo {
  private static FileSystem initFS() throws IOException {
      Configuration conf = new Configuration();
      // For more information on COSN configuration items, visit https://intl.cloud.tencent.com/document/product/436/6884?from_cn_redirect=1#hadoop-.E9.85.8D.E7.BD.AE
      // The following configuration items are required
      conf.set("fs.cosn.impl", "org.apache.hadoop.fs.CosFileSystem");
      conf.set("fs.AbstractFileSystem.cosn.impl", "org.apache.hadoop.fs.CosN");
      conf.set("fs.cosn.tmp.dir", "/tmp/hadoop_cos");
      conf.set("fs.cosn.bucket.region", "ap-guangzhou");
      conf.set("fs.cosn.userinfo.secretId", "AKXXXXXXXXXXXXXXXXX");
      conf.set("fs.cosn.userinfo.secretKey", "XXXXXXXXXXXXXXXXXX");
      conf.set("fs.ofs.user.appid", "XXXXXXXXXXX");
      // For more information on other configuration items, visit https://intl.cloud.tencent.com/document/product/436/6884?from_cn_redirect=1#hadoop-.E9.85.8D.E7.BD.AE
      // Whether to enable CRC-64 checksum. It is disabled by default, meaning that you can’t run the `hadoop fs -checksum` command to obtain the CRC-64 checksum of a file.
      conf.set("fs.cosn.crc64.checksum.enabled", "true");
      String cosnUrl = "cosn://f4mxxxxxxxx-125xxxxxxx";
      return FileSystem.get(URI.create(cosnUrl), conf);
  }
   private static void mkdir(FileSystem fs, Path filePath) throws IOException {
      fs.mkdirs(filePath);
  }
   private static void createFile(FileSystem fs, Path filePath) throws IOException {
      // Create a file (if it already exists, it will be overwritten)
      // if the parent dir does not exist, fs will create it!
      FSDataOutputStream out = fs.create(filePath, true);
      try {
          // Write a file
          String content = "test write file";
          out.write(content.getBytes());
      } finally {
          IOUtils.closeQuietly(out);
      }
  }
   private static void readFile(FileSystem fs, Path filePath) throws IOException {
      FSDataInputStream in = fs.open(filePath);
      try {
          byte[] buf = new byte[4096];
          int readLen = -1;
          do {
              readLen = in.read(buf);
          } while (readLen >= 0);
      } finally {
          IOUtils.closeQuietly(in);
      }
  }
   private static void queryFileOrDirStatus(FileSystem fs, Path path) throws IOException {
      FileStatus fileStatus = fs.getFileStatus(path);
      if (fileStatus.isDirectory()) {
          System.out.printf("path %s is dir\n", path);
          return;
      }
      long fileLen = fileStatus.getLen();
      long accessTime = fileStatus.getAccessTime();
      long modifyTime = fileStatus.getModificationTime();
      String owner = fileStatus.getOwner();
      String group = fileStatus.getGroup();
       System.out.printf("path %s is file, fileLen: %d, accessTime: %d, modifyTime: %d, owner: %s, group: %s\n",
              path, fileLen, accessTime, modifyTime, owner, group);
  }
  
  private static void getFileCheckSum(FileSystem fs, Path path) throws IOException {
      FileChecksum checksum = fs.getFileChecksum(path);
      System.out.printf("path %s, checkSumType: %s, checkSumCrcVal: %d\n",
              path, checksum.getAlgorithmName(), ByteBuffer.wrap(checksum.getBytes()).getInt());
  }
   private static void copyFileFromLocal(FileSystem fs, Path cosnPath, Path localPath) throws IOException {
      fs.copyFromLocalFile(localPath, cosnPath);
  }
   private static void copyFileToLocal(FileSystem fs, Path cosnPath, Path localPath) throws IOException {
      fs.copyToLocalFile(cosnPath, localPath);
  }
   private static void renamePath(FileSystem fs, Path oldPath, Path newPath) throws IOException {
      fs.rename(oldPath, newPath);
  }
   private static void listDirPath(FileSystem fs, Path dirPath) throws IOException {
      FileStatus[] dirMemberArray = fs.listStatus(dirPath);
       for (FileStatus dirMember : dirMemberArray) {
          System.out.printf("dirMember path %s, fileLen: %d\n", dirMember.getPath(), dirMember.getLen());
      }
  }
   // The recursive deletion flag is used to delete directories
  // If recursion is `false` and `dir` is not empty, the operation will fail
  private static void deleteFileOrDir(FileSystem fs, Path path, boolean recursive) throws IOException {
      fs.delete(path, recursive);
  }
   private static void closeFileSystem(FileSystem fs) throws IOException {
      fs.close();
  }
   public static void main(String[] args) throws IOException {
      // Initialize a file
      FileSystem fs = initFS();
       // Create a file
      Path cosnFilePath = new Path("/folder/exampleobject.txt");
      createFile(fs, cosnFilePath);
       // Read a file
      readFile(fs, cosnFilePath);
       // Query a file or directory
      queryFileOrDirStatus(fs, cosnFilePath);
       // Get a file checksum
      getFileCheckSum(fs, cosnFilePath);
       // Copy a file from the local system
      Path localFilePath = new Path("file:///home/hadoop/ofs_demo/data/exampleobject.txt");
      copyFileFromLocal(fs, cosnFilePath, localFilePath);
       // Download a file to the local file system
      Path localDownFilePath = new Path("file:///home/hadoop/ofs_demo/data/exampleobject.txt");
      copyFileToLocal(fs, cosnFilePath, localDownFilePath);
       listDirPath(fs, cosnFilePath);
      // Rename
      mkdir(fs, new Path("/doc"));
      Path newPath = new Path("/doc/example.txt");
      renamePath(fs, cosnFilePath, newPath);
       // Delete a file
      deleteFileOrDir(fs, newPath, false);
       // Create a directory
      Path dirPath = new Path("/folder");
      mkdir(fs, dirPath);
       // Create a file in a directory
      Path subFilePath = new Path("/folder/exampleobject.txt");
      createFile(fs, subFilePath);
       // List directories
      listDirPath(fs, dirPath);
       // Delete a directory
      deleteFileOrDir(fs, dirPath, true);
      deleteFileOrDir(fs, new Path("/doc"), true);
       // Close a file system
      closeFileSystem(fs);
  }
}

FAQs

If you have any questions about Hadoop-COS, see Hadoop.