How to Improve Classification Task Time - Parallel Classification

(Doc ID 2405895.1)

Last updated on MAY 31, 2018

Applies to:

Oracle Knowledge - Version 8.6 and later
Information in this document applies to any platform.

Goal

Classification always classifies all collections.
 
Generally the biggest problem with classification is memory.  Classification does a lot of its work in memory.  When the collections size get really large this is one good reason to divide collections if they are too large.  How to improve crawl performance?(Doc ID 1039026.1)  Also tuning the java GC will help with performance.  If you see Allocation Failures then GC tuning should be performed.  How to Resolve GC Allocation Failure Messages(Doc ID 2396339.1)

Certain facets take a long time to classify and slow down the classification.  These facets may be a business requirement.
Fixing Out Of Memory Error During Classification with CMS-GUID enabled(Doc ID 1041031.1)
 
After you look at these considerations you can also consider running parallel classification.  Parallel Classification is introduced in the 8.6.1 patch readme.  If you do not already have workclients you will need to add workclients to your system.  These are separate indexer type instances that perform some indexing task processes.

Content Processing - Recommended Environment Setup for Indexer and Workclients(Doc ID 1041144.1)

Using Remote Installer Install Staging/ Production or to Add A Runtime Or WorkClient Instance To Existing Search Configurations (Doc ID 1622499.1)

Solution

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms