• MapReduce Service

  1. Help Center
  2. MapReduce Service
  3. User Guide
  4. FAQs
  5. How Do I Prepare a Data Source for MRS?

How Do I Prepare a Data Source for MRS?

MRS can process data in both OBS and HDFS. Before using MRS to analyze data, you are required to prepare the data.

  1. Upload local data to OBS.
    1. Log in to the OBS management console.
    2. Create a userdata bucket, and then create the programinputoutput, and log folders in the userdata bucket.
      1. Click Create Bucket to create a userdata bucket.
      2. In the userdata bucket, click Create Folder to create the programinputoutput, and log folders.
    3. Upload local data to the userdata bucket.
      1. Go to the program folder, and click  to select a user program.
      2. Click Upload.
      3. Repeat preceding steps to upload the data files to the input folder.
  2. Import OBS data to HDFS.

    This function is available only when Kerberos authentication is disabled and the cluster is running properly.

    1. Log in to the MRS management console.
    2. Go to the File Management page and select HDFS File List.
    3. Click the data storage directory, for example, bd_app1.

      bd_app1 is just an example. The storage directory can be any directory on the page. You can create a directory by clicking Create Folder.

    4. Click Import Data, and click Browse to configure the paths of HDFS and OBS, as shown in Figure 1.
      Figure 1 Importing files
    5. Click OK.

      You can view the file upload progress in File Operation Record.