In contrast to monitoring tools that simply present measurement data, Performance Manager contextualizes monitor results with health rates. Health rates enable analysts who have little familiarity with monitored applications to readily evaluate monitor results. They also assist experienced analysts in pinpointing relevant measurements.
The primary challenge associated with interpreting monitor reports is that when you're not familiar with their format, you need to spend time analyzing their meaning. Even when you are able to efficiently interpret a report format associated with a certain project, when you view a different report format designed for a different project you must once again evaluate what is relevant and what can be ignored. In response to this challenge, Performance Manager converts all of its measurement data into percentage rates. All relevant data is thereby converted into abstract percentage values that are aggregated into single high-level values, or health rates, that reflect overall project readiness.
The benefit of health rates is that they offer analysts a short cut for evaluating project health and directing development efforts. If the overall health rate of a project is solid, there is no need for further analysis. Health rates have values between "0" and "100" (0 being the worst, 100 being the best) and are independent of projects, the amount of data analyzed, and the frequency of individual measurements.
Because high-level health rates are the aggregate of low-level health rates, analysts have the option of reversing rate calculations to determine how specific low-level rates influence overall rates. Such causal analysis can be used to "drilldown" to specific low-level data that negatively affects overall rates, thereby pinpointing the system components that have a negative impact on system health. All the while, measurements that fall within acceptable ranges are ignored.
Because low-level health rates reveal the fitness of actual measurement values, analysts don't need to understand the significance of measurement values themselves. For example, without familiarity with a certain monitored application, it would not be readily apparent whether a business transaction that takes 15 seconds is fast or slow. A health rate of "95%" however is readily understood to be a healthy rate.
There are several aspects of monitored results. The first aspect is result type (response time, error message, counter, etc.). Each monitored result is also associated with a certain transaction, which runs on a specific location. Once rates are calculated they are aggregated into result type, and ultimately used to generate the three health-dimension rates: Availability, Accuracy, and Performance.
The first step in evaluating poor health rates is to examine the associated health dimensions. Say for example that a certain monitor or agent encounters problems with a server under test. In such an instance, only the relevant data related to the server experiencing problems would be presented. In this way, Performance Manager simplifies the process of identifying the causes of detected problems.
To calculate how expected measurement values correlate with the health rate scale of 0-100 (without the input of an expert who is familiar with the capabilities of the system under test), one has to rely on comparisons of past measurement values. In the same way that an athlete estimates his fitness by comparing his recent achievements to past performance, Performance Manager compares measured values to measurement values recorded in the past. This method evaluates what is possible, but it does not offer an indication of maximum performance potential. Load testing and benchmarking are better suited for such assessments. Note that with this calculation approach, it must be assumed that production systems are performing well and that variations from benchmark rates warrant analyst scrutiny.