Migrating to Cloudera on CloudPDF version

Creating collections for migration

After scanning the source cluster, you can use collections to sort the datasets that need to be migrated to the destination cluster. Collections can also be useful to organize different types of data, workflows and applications before migration.

  1. Click Start Mapping or Collections.
  2. Click Hive Tables, HDFS Location, Hive SQL, HBase Tables, Oozie Job Definitions or Spark Applications based on which items you would like to add to a collections.
    The different windows show the results of the scanning. For example, the Hive Tables display all the existing tables in Hive on the source cluster.
  3. Select the items that you want to add to a collection, and click Add to collection.
    The selected items are added to the Default collection, and the Default label is assigned to the selected items.
    You have the option to create more collections beside the Default collection.
  4. Click next to Collections, and customize the collection by providing a Label name and selecting a Color. Click Create.
    Select more items from the scanning results, and add it to the newly created collection.
    You can manage the created collections by clicking on to open the collections menu. You can review the labeled results in a collections using View Items, and modify or delete the created collection using Edit and Remove. Removing a collection does not affect the items on the source cluster, only the labelling is deleted.
The datasets are labeled for migration.
Register the destination cluster to which the datasets and workflows are going to be migrated.

We want your opinion

How can we improve this page?

What kind of feedback do you have?