Providing Open Architecture High Availability Solutions

The scope of this document includes only functions and capabilities specifically related to the key

parts of an HA system. These parts are:

1. Redundant hardware and software components

2. Methods for storing information about the components and their relationships

3. Methods for managing faults, from detection through recovery and repair

4. Methods for replacing and upgrading components and updating the stored information

5. Dynamic reconfiguration of the system

Functions normally needed to maintain a secure and well-managed system, such as load

management, administration, security (physical and logical), and basic component management,

will be included only when needed to explain the main subject.

Topics which are related, but not central to, HA system design will be covered by reference. These

topics include such items as:

• Performance management

• Redundant application databases

There are problems that can occur which will make a system unavailable that have more to do with

general system practices (administration, security) then they do with HA system design. As such,

solving the following problems for the general case is outside the scope of this document:

• Malicious attacks (e.g., “denial of service” network attacks)

• Survivability (from earthquakes, fires, floods, etc.)

• Network congestion

• System overload