5.2. Retrieving RDX Packages#

Data czars retrieve RDX packages from AWS. The procedures for downloading RDX data packages and extracting their contents is similar to the procedures you follow for your organization’s data packages.

The topics in this section describe the differences between RDX packages and organizational data packages.

For more information about downloading and extracting your organizational data packages, see Data Delivered in Data Packages.

5.2.1. Identifying an RDX Package#

When researchers propose projects that require RDX data, they request data either for specific courses or for courses that meet specified criteria. EdX identifies courses that meet these specifications and, on approval, assigns an RDX request ID to the project.

EdX provides a single RDX data package for each approved project. The RDX package contains all of the data files, after the data obfuscation procedures are applied, for all of the courses in the project. On AWS, data czars can identify the RDX packages by request ID.

5.2.2. Locating RDX Data on Amazon S3#

EdX uploads RDX data packages to an Amazon S3 s3://edx-course-data/{org}/rdx/{requestID} folder. This folder contains one {org}_{course}_{run}.tar.gz.gpg file for each of the requested courses.

The {org}_{course}_{run}.tar.gz.gpg file contains data and event files for the requested time period in the course’s history.

5.2.3. Download an RDX Package from Amazon S3#

Only the data czar at a participating partner institution can download an RDX data package.

To download an RDX data package from the Amazon S3 storage service, follow these steps.

Note

If you are using a third party tool to connect to Amazon S3, you might not be able to navigate directly from the bucket that contains your own organization’s data to the bucket for RDX data. You might need to disconnect from Amazon S3 and then reconnect to the RDX destination.

  1. Use the AWS Command Line Interface or a third party tool to connect to the s3://edx-course-data/ folder on Amazon S3.

    For information about providing your credentials to connect to Amazon S3, see Access Amazon S3.

  2. Navigate to the RDX folder for a specific research project. The RDX data package is located in the s3://edx-course-data/{org}/rdx/{requestID} folder.

  3. Download the compressed, encrypted {org}_{course}_{run}.tar.gz.gpg file for each of the courses in the project.

5.2.4. Extract the {org}_{course}_{run}.tar.gz.gpg File#

After you download each of the {org}_{course}_{run}.tar.gz.gpg files for an RDX research project, you complete these tasks. You repeat this process for each {org}_{course}_{run}.tar.gz.gpg file in the RDX package.

  1. Use your private key to decrypt the file. The result is a file named {org}_{course}_{run}.tar.gz. See Decrypt an Encrypted File.

  2. Use the gunzip command to expand the .gz file. The result is a file named {org}_{course}_{run}.tar.

  3. Extract the .tar file. The result of this decryption and extraction process is a metadata_file.json file and a set of folders that contain the event and state data for the requested time period. The file and folder hierarchy follows this pattern.

    metadata_file.json
  /events
    {org}_{course}_{run}-events-CCYY-MM-dd.log.gz
    ...
  /state
    /CCYY-MM-dd
      {org}_{course}_{run}-{suffix}
      ...

    The metadata_file.json file provides information about the version of the edX analytics pipeline that was in use when edX obfuscated PII and other sensitive values in the package.