Analyzing benchmark results
In this section, we will analyze the benchmark results and identify how the 120,000 requests affected our application. In creating web application benchmarks, there are typically two things we are usually interested in:
- How many requests can our application handle at a time?
- For how long is each request being processed by our application?
These two low-level web-performance metrics can easily translate to the business implications of our application. For example, how many customers are using our application? Another one is, how are they perceiving the responsiveness of our application from a user-experience perspective? We can correlate secondary metrics in our application such as CPU, memory, and network to determine our system capacity.
Viewing the results of JMeter runs
Several listener elements of JMeter have features that render graphs. Enabling this when running the benchmark is useful when developing the test plan. But the time taken by the UI to render the...
