Because a portal service is normally critical to conducting business, it must be maintained with minimal disruption and downtime.
Common servicing operations include database backups, replacement of applications and system software, upgrades, and other maintenance. Analyzing the solution's requirements for such servicing, and techniques to facilitate the servicing, should be a priority when you design the system.
For example, on an intranet-oriented portal, users are generally most active during the 8:00 a.m. to 5:00 p.m. working hours. This means all system servicing operations can be done after hours. However, if your organization is geographically distributed over multiple time zones, or users need 24–by-7 access to the system, there is no servicing window. Instead, the system needs to be designed so that all maintenance operations can be done with the system in operation or having little impact on the system's availability. In addition to an appropriate deployment architecture, it is necessary to have well-defined and tested operational procedures that ensure minimum downtime.
The Portal Service on Application Server Cluster reference configuration is designed to maximize serviceability, both with respect to scaling the portal service and upgrading software components in the configuration.