CTFL-PT practice question 9 of 10

You have been asked to test the performance of a software and system update to a North American mobile trivia quiz game application. The quiz app is…

You have been asked to test the performance of a software and system update to a North American mobile trivia quiz game application. The quiz app is available on major app stores serving English- speaking users. The number of concurrent users planned for is 5 million Winning quiz participants are promised a share of an allocated cash prize Once participants start playing, their quiz sessions must not be lost especially if all previous questions were answered correctly. The application owner wants to ensure that 99% of quiz sessions are maintained (up to 4 95 million sessions) until the last answer has been successfully recorded In order to avoid cooperation among quiz participants, the answers to the multiple choice questions are listed randomly Confidence in the game's fairness requires that a registered user originate from only one quiz client device Every user must respond over the same short window of time in order to avoid external assistance and learning the official answer before responding. Only participants who respond correctly may answer the next question Any activity that appears to violate expected behavior of a rule-following participant will cause the client session to be rejected. You are currently preparing test scripts to function as a quiz participant. What input data type is most critical for the test script to incorporate to ensure that the concurrency objective will be properly tested?

Answer options

A: In order to property test the concurrency objective, the test script must answer each trivia question with the same answer as every other virtual user

B: In order to property test the concurrency objective, the test script must provide a random password to the application's login system

C: In order to properly test the concurrency objective, the test script must present a unique user name to the application's login system Correct answer

D: In order to properly test the concurrency objectives the test script must supply the system provided unique session ID with each answer provided

Correct Answer:

C. In order to properly test the concurrency objective, the test script must present a unique user name to the application's login system

Explanation

The primary objective is to test the system's performance with 5 million concurrent users. To accurately simulate this scenario, each virtual user in the performance test must represent a unique, real-world participant. This is achieved by ensuring each virtual user logs in with a unique set of credentials. The most critical piece of input data to establish a unique user session is the username. Using unique usernames ensures that the system creates and manages 5 million distinct sessions, thereby properly testing its ability to handle the specified concurrent load as described in the requirements.

Why Incorrect

A. Using the same answer for all users does not test concurrency realistically; it only tests a single, non-representative data path under load.
B. A random password, when used with a pre-existing username, would cause authentication to fail, preventing the script from establishing a user session and contributing to the concurrency test.
D. The session ID is typically generated by the server after a successful login. It is an output from the system, not an initial input from the test script to establish the concurrent session itself.

References

1. ISTQB® Certified Tester Foundation Level Performance Testing Syllabus Version 1.0 (2018), Section 3.3.2, Test Data, Page 26: "The test data used for a performance test should be as realistic as possible... For example, if the system under test is a multi-user system, the test data should include a sufficient number of unique user accounts to support the planned number of concurrent users." This directly supports the need for unique usernames to simulate the concurrency objective.

2. Jain, R. (1991). The Art of Computer Systems Performance Analysis: Techniques for Experimental Design, Measurement, Simulation, and Modeling. Wiley. Chapter 3, "Workloads," emphasizes the importance of a representative workload for accurate performance testing. A representative workload for a multi-user application requires simulating distinct users, which necessitates unique user data.

3. Jiang, Z. M., & Hassan, A. E. (2015). A survey on load testing of large-scale software systems. IEEE Transactions on Software Engineering, 41(11), 1091-1118. (DOI: 10.1109/TSE.2015.2445201). This survey discusses workload modeling, where user behavior and characteristics are simulated. Simulating 5 million concurrent users fundamentally requires 5 million unique user profiles or identities to be part of the test data.

Keep practicing

Take the free CTFL-PT practice test

Ten exam-style questions with answers and explanations, plus the exam facts and study guides.

Start the free practice test