Why should the Response Time Composition Chart in AppResponse NOT be used on All Traffic or other large traffic Groups? (Choose all that apply.)

The Response Time Composition Chart in AppResponse displays average performance metrics for the selected traffic group. When applied to a broad, heterogeneous group like "All Traffic," performance data from numerous different applications, clients, and servers is aggregated and averaged. This statistical averaging masks the true performance of individual components. Consequently, high-latency events or poorly performing applications (performance anomalies) are obscured or "smoothed out" by the large volume of normal, low-latency traffic. This makes the chart misleading for troubleshooting, as the root cause of a specific issue cannot be isolated. Effective analysis requires using the chart on narrowly defined, specific application groups.

D. The chart is a core feature of Application Stream Analysis (ASA) and applies to any defined TCP-based application group, not exclusively to Web Transaction Analysis (WTA) traffic.

E. This chart is a high-level dashboard component for initial analysis. Packet Analyzer Plus is a separate tool for deep-dive packet inspection, typically used after a problem is identified.

1. Riverbed AppResponse 11.12 User's Guide, Chapter 4: Application Stream Analysis, Section: "Application Stream Analysis Dashboards," Page 101.

This section describes the "Response Time" widget (the composition chart), stating it "Shows the average response time for the selected application..." This confirms that the chart uses averages, which is the basis for the problems described in options A, B, and C. The section also places this chart within the general Application Stream Analysis (ASA) functionality, refuting the claim in option D that it is only for WTA.

2. Riverbed AppResponse 11.12 User's Guide, Chapter 4: Application Stream Analysis, Section: "Application Stream Analysis Workflow," Page 100.

This section outlines the intended workflow, which starts with high-level dashboards and then allows for "drill down to see more detailed information" or to "launch Packet Analysis for deep packet-level analysis." This refutes the claim in option E that using Packet Analyzer is a prerequisite.

3. Jain, R. (1991). The Art of Computer Systems Performance Analysis. John Wiley & Sons. Chapter 10: "The Art of Data Presentation."

This foundational academic text on performance analysis explains that while summary statistics like averages are useful, they can also hide significant details. Section 10.2 warns that "the first thing you should determine is whether the variability is small or large. If it is large, the mean and variance may not be sufficient." Applying a single average to a large, diverse group like "All Traffic" is a classic example of improperly representing a system with high variability, which supports the reasoning in options A, B, and C.