3 key capacity advisor concepts, Capacity planning goals, Quality of service – HP Matrix Operating Environment Software User Manual

Page 23: Utilization monitor, calculator, and simulator, Data handling for virtual machines

Advertising
background image

3 Key Capacity Advisor concepts

Capacity planning goals

Capacity planners are driven by multiple, sometimes conflicting, goals:

Avoiding surprises

Planning for the future

Maintaining quality of service (see

“Quality of service ” (page 23)

)

Optimizing resource utilization

Meeting internal and external security requirements

Reducing costs

Capacity Advisor can help lessen the conflicts among goals by buttressing decisions with better
information and more robust models. Effective use of Capacity Advisor can help you to double
resource utilization in many situations.

The most important step in accomplishing capacity planning goals is to clearly understand the
goals and the relationship among them. For example, optimizing response time may preclude
trying to achieve the most efficient resource utilization, while optimizing resource utilization can
compromise certain types of security. It is critical to understand the requirements as well as their
priorities to provide best overall quality of service.

Quality of service

Typical aspects of quality of service include (but are not limited to):

Availability

Accessibility

Integrity

Performance

Reliability

Regulatory requirements

Security

Quality of service is frequently defined in service level agreements between organizations.

Utilization monitor, calculator, and simulator

At a fundamental level, Capacity Advisor collects data from utilization monitoring daemons on
systems and workloads. Using this data, Capacity Advisor can report on utilization of memory,
core processing units, I/O bandwidth, and power. It can calculate what resource utilization would
look like if the load were increased or decreased. Further, it can simulate what resource utilization
might look like if loads are combined. This goes beyond simple addition of maximum loads to
provide a dynamic addition of components based on the time sequence of utilization measurements
taken from the actual loads running on real systems (see

“Peaks and sums” (page 25)

).

Data handling for virtual machines

Data collected by Capacity Advisor for use in a profile viewer and historical utilization reports can
vary from data reported by tools that run inside a virtual machine. These tools include HP-UX and
Linux commands such as top, ps, and sar, and on Microsoft Windows, the task manager or
perfmon

.

This data can vary in a couple of ways. For example:

Capacity planning goals

23

Advertising