File System Application Requirements

The following tables include the minimal node requirements for building and running file system applications.

Node Requirement Method(s) to Meet Requirement
A connection to the Data Fabric cluster.
Select one of the following options:
  • Install and configure the Data Fabric client.
  • Install the PACC and run an application container.
  • Use the file system JAR to connect to the cluster.

For more information, see Connect to the Cluster.

Include hadoop-common libraries as a dependency.
When you compile the application, use the Maven Repository to determine the dependencies.The POM file should include the Data Fabric Repository and the hadoop-common dependency:
<repositories> 
   <repository> <id>mapr-releases</id>
    <url>https://repository.mapr.com/maven/</url>
<snapshots><enabled>false</enabled></snapshots>   <releases><enabled>true</enabled></releases>
</repository>
   </repositories>

<dependencies>
       <dependency>
       <groupId>org.apache.hadoop</groupId>
       <artifactId>hadoop-common</artifactId>
       <version>${hadoop.version}</version>
       </dependency>
</dependencies>

When you run the application, include the following in the application’s classpath: `hadoop classpath`

For more information, see Compiling and Running a Java Application.

NOTE
When you develop a Java application, you can use a dependency management tool such as Maven to compile your application. However, it is recommended that you do the following instead:
  1. Compile the Java application without including dependencies
  2. Specify the required classpath when you submit the application to the cluster
If you choose to bundle the JAR file, and there is a mismatch between the bundled JAR file and the version that your Data Fabric cluster expects, this can result in failures. The failures differ depending on the version of Data Fabric you are using. For more information, see Using the File System JAR to Connect to the Cluster.
Node Requirement Method(s) to Meet Requirement
A connection to the Data Fabric cluster.
Select one of the following options:
  • Install and configure the Data Fabric client.
  • Install the PACC and run an application container.

For more information, see Connect to the Cluster.

Include the libhdfs libraries and Data Fabric libraries when you compile the application. The Data Fabric libraries are available in the following location: /opt/mapr/lib

Link to the libhdfs libraries in the following location: MAPR_HOME/hadoop/hadoop-2.x/

For more information, see Compiling and Running C Applications on File System Clients