Load Balancing - User requests are distributed across nodes for performance and availability.
High availability - Ability for one node to go offline without significantly impacting users.
Scalability - Additional capacity may be added as needed.