A large AEM enterprise site is implementing authentication and requires a true optimal load balancing across the site's multi AEM publish instances. Which approach should an Architect take to meet…

Question

Options

• AConfigure Sticky Connections
• BConfigure the Dispatcher to run with a Dedicated System Usei
• CEnable Encapsulated Token Option
• DEnable Round Robin processing in Dispatcher configuration

• A
  3% (2)
• B
  6% (4)
• C
  13% (9)
• D
  79% (56)

Explanation

Round Robin processing in the Dispatcher (/lbMethod "byrequests" or similar) distributes incoming requests evenly and sequentially across all AEM publish instances, achieving true optimal load balancing without favoring any single node - making D the correct architectural choice.

Why the distractors fail:

• A (Sticky Connections): Sticky connections (session affinity) pin a user to the same publish instance for their session, which defeats true load balancing by creating uneven server loads - ironically, it's often used instead of true load balancing when session state can't be shared.
• B (Dedicated System User): This is a security best practice for service accounts accessing the JCR repository; it has no relation to load balancing or request distribution.
• C (Encapsulated Token): The Encapsulated Token feature makes authentication tokens self-contained so they're valid on any publish instance (enabling stateless auth), but it's an authentication mechanism, not the load balancing configuration itself. It's a complement to Round Robin, not a replacement.

Memory tip: Think "Sticky = stuck" - sticky connections stick a user to one server and break true load balancing. Round Robin rotates requests, which is what "true optimal" distribution means. On the exam, any option involving "sticky" is the anti-pattern to load balancing.

No community discussion yet for this question.