Availability management starts with configuring the various hardware and software components with their individual reliability characteristics into a system that meets the availability requirements of the end user. It covers all the tasks associated with the delivery of service to the end users.
You might expect that Linux on the mainframe enables you to provide a higher available service at reduced cost when compared to alternative server farm configurations. In this chapter, we will cover some of the areas where we believe you will find key leverage points for this value potential.
We will regard a server as available if the end user can perform the desired functions. This definition of availability spans a wide range of failure points, from power to the hardware, the processors running, the disks spinning, and the network running to the operating system, the middleware, and applications all running and functioning together.
We will discuss all tasks listed in Figure 13-1 to some degree, but we will primarily focus on the following three availability tasks because they potentially have some unique value for you in a Linux-on-the-mainframe environment:
Figure 13-1. Availability management tasks
We will focus on how Linux, the mainframe, and z/VM can all contribute to the management tasks of keeping the system available according to your policies. We will explore questions such as: