Following are the Performance Testing Interview Questions Set #10.
Q. 46 What are the performance bottlenecks that you found in the projects you were working? What are the recommendations made to overcome those issues?
Ans: The following are the bottlenecks found in the application during performance testing:
- The bottleneck in the JavaScript and CSS request increases the response time of the application webpage.
- While logging into the application which includes database-related transactions; it seemed very slow, so I started my investigation by looking into the tables in the database. I found that indexing was not present in the tables. Once the indexing is attached to the tables readability becomes very fast and response time improves.
Q. 47 What do you monitor during performance test execution?
Ans: During performance test execution of an application, we always look for both client-side as well as server-side metrics.
Client-side metrics include transaction response time, throughput, hits per second, Running Users etc. and server-side metrics include CPU Utilization, Memory consumption, garbage collector performance etc.
Q. 48 How to extract server data for test execution and how to analyze that?
Ans: Every time post-test execution, we need to extract the data from the server which is the CPU, Memory, Garbage collector, and Heap Memory which may be in graphical representation. We need third-party tools like Dynatrace, JavaMelody, Jvisual VM etc. to get these metrics.
Q. 49 What is the SLA (service level agreements)?
Ans: You can define service level agreements (SLAs) which measure performance test goals over time intervals during a test run, or over a whole performance test run. This is a mutual agreement between the client and the performance testing team to define accurate pass/fail criteria of the scenario.
Examples are:
- Transaction Response Time (Average) per time interval
- Errors per Second per time interval
- Total Hits per run
- Average Hits (hits/second) per run
- Total Throughput (bytes) per run
- Average Throughput (bytes/second) per run etc.
Q. 50 How to identify performance bottlenecks?
Ans: In order to find out the performance bottleneck, it is important to identify the root cause of the issue. Firstly analyse the performance testing graph and filter out the time window when the issue occurred. Secondly, correlate the client-side metric graph with the server-side metric graph and analyse the server behaviour at that particular time. After that investigate the server logs and DB logs and find out the root cause. The application performance monitoring (management) tools make the root cause identification process easier and quicker.