Processing

Use Relativity’s processing feature to ingest raw data directly into your workspace for eventual search, review, and production without the need for an external tool. You can use the various processing objects to create custom processing jobs that handle a wide array of information.

Some of the primary goals of processing are to:

  • Discern, at an item level, exactly what data is found in a certain source.
  • Record all item-level metadata as it existed prior to processing.
  • Enable defensible reduction of data by selecting only items that are appropriate to move forward to review.

Note: Processing doesn't perform language identification. For information on how to perform language identification using Analytics, see Language identification.

To gain control over more complex processing jobs on a granular level, you can use the Relativity Processing Console desktop application. For more information, see the Relativity Processing Console.

This page contains the following information:

See these related recipes:

Application version considerations

All the content in this section and its related pages correspond to the latest version of the Processing application, which is updated on a monthly basis with the release of each patch of Relativity 9.6.

If the processing components in your environment don't match the descriptions in this content exactly, it may be because you're using an older version of the Processing application. To get the newest version of the Processing application, upgrade to the latest product update of Relativity 9.6.

For a list of changes made to processing per monthly product update, see the Release Notes.

Basic processing workflow

The following graphic and corresponding steps depict a typical processing workflow that uses all available processing objects and phases. Note that each user's workflow may vary. You may not be required to follow all of these steps for every processing job you run.

  1. Create new custodian entries in the Entities tab to correspond with the individuals who hold the files you intend to process. Eventually, you'll associate each custodian with a data source.
  2. Create new password bank entries to unlock any password-protected files you intend to process.
  3. Create a new processing profile to specify the settings you want the processing engine to refer to when completing your processing jobs. You'll select this profile when creating a processing set.
  4. Create a new processing set that uses the processing profile you created. Then, add processing data sources to your saved processing set to specify the custodians and locations of the files you want to process.
  5. Inventory the files located in the data sources associated with your processing set. This is optional.
  6. Apply filters to the inventoried files to further narrow down the data set before discovery. This is optional.
  7. Run reports to gauge how much you've narrowed down your data set to this point. You can also run reports after discovery and/or publish.
  8. Discover the inventoried and filtered files. Then, publish the discovered files to your workspace so that you can review and eventually produce them.
  9. View, ignore, or retry any errors that occurred during any phase of the processing job. If needed republish the files.
  10. Search, review, and produce the documents that you published to your workspace.

Logging for processing

The logging framework enables you to efficiently gather runtime diagnostic information. You can use logging for troubleshooting application problems when you need a very granular level of detail, for example, when working with a Relativity Support representative.

Relativity system components that can log messages are identified based on the system-subsystem-application designation. When troubleshooting, use the system-subsystem-application matrix to configure logging to target a specific Relativity component, such as the Processing application.

Note: It's recommended that you not set your logging to verbose when publishing documents to a workspace, as doing so can cause your worker to run out of resources (CPU, RAM, disk usage, etc.), which then causes your publish job to cease entirely. If you need to use verbose logging in order to collect detailed logs, do so for short periods of time only (under 5 minutes) and have a developer on hand to troubleshoot if any issues occur.

For more information, see Logging and Logging system-subsystem-application matrix