During an optimization session, a client using AEM 6.5 mentions that they face challenges with their AEM Author setup. The client has significant repository growth and can see an increased amount of…

Question

Options

• ALeverage workflow offloading for all workflows
• BConfigure the Maximum Parallel Jobs
• CIncrease the maxFileSize of the TAR segmentstore
• DLeverage Transient Workflows where possible
• EExecute an offline tar compaction twice a day

• A
  3% (1)
• B
  70% (21)
• C
  20% (6)
• E
  7% (2)

Explanation

B and D directly address the two symptoms described - slow workflows and repository growth - without requiring downtime or architectural overhaul. Configuring Maximum Parallel Jobs (B) tunes how many workflow instances the Granite Workflow Engine processes simultaneously; the default may be too low for the editor workload, so investigating this setting can improve throughput. Transient Workflows (D) don't persist workflow instance data in the JCR repository, which directly reduces repository growth and TAR file accumulation - the root cause of the disk expansion described.

A is wrong because offloading all workflows is overly broad and adds topology complexity; offloading is a targeted solution for specific long-running jobs, not a general optimization. C is wrong because increasing maxFileSize just makes individual TAR segments larger - it doesn't reduce the volume of data being written or improve performance in any meaningful way. E is wrong because offline compaction requires AEM to be shut down, making it impractical to run twice a day on an active Author instance; online Revision Cleanup is the correct recurring approach.

Memory tip: Think "T·P" - Transient + Parallel. Transient = less data written (smaller repo), Parallel = more jobs processed (faster workflows). Both are non-disruptive Author-side tuning options.

No community discussion yet for this question.