Top 10 Performance Testing Interview Questions

Article about Top 10 Performance Testing Interview Questions

Published by

Sep 15, 2021


In case you are intending to begin your vocation as a performance tester, Below are a couple of the most regularly posed Performance Testing Interview Inquiries. These inquiries were posed in some top MNCs like Oracle, TCS, Accenture, Infosys, Wipro, HCL, Capgemini, and so forth 


Q1. What are the Various Sorts of Performance testing? 


Ans: There are six significant sorts of Performance Testing: 


1. Load Testing: The motivation behind load testing is to recognize application performance under a specific fixed load. 


2. Stress Testing: The objective of stress testing is to distinguish the breakpoint of an application. 


3. Endurance Testing: Endurance test is executed throughout a significant period ( Eg: 4 hrs, 8hrs, 1 day, and so on) to recognize the issues like memory spill in an application. 


4.Spike Testing: The motivation behind spike testing is to distinguish the application performance under an unexpected increment or decline in load. 


5. Failover Testing: Failover testing is a method to check the framework's capacity to give additional assets and the capacity to move to back-up frameworks during the framework's disappointment because of either reason. 


6.Volume Testing: Volume testing is done to investigate the framework performance by expanding the volume of information in the data set. 


Q2. What is the contrast between Stress and Endurance testing? 


This is quite possibly the most asked performance testing interview inquiry. 


Ans: A stress test is executed to recognize the breakpoint (Maximum measure of load an application can maintain) of an application anyway endurance test is executed throughout an extensive period (4hr, 8hr, 1 day, and so on) to distinguish the issues like memory spill. 


Q3. How would you characterize the term for an endurance test? 


Ans: The span of an endurance test is characterized dependent on the utilization example of end-users. 


1. On the off chance that the end-users will utilize the application under test (AUT) during the standard available time just, then, at that point we can set the endurance test span as 8 hours. 


2. On the off chance that the application under test (AUT) will be utilized all around the world, we can set the length of an endurance test as at some point or even multi-week. 


Q4. What are capacity testing and capacity arranging? 


Ans: The capacity test is a test to decide the number of users your application can support with the given equipment. These sorts of tests are for the most part executed to recognize equipment constraints. 


Capacity test results are useful in capacity arranging. 


For instance, We executed a capacity test on an application worker having 8GB RAM and saw that the worker can support a limit of 100 client's load. 


In light of the above outcomes, If we are wanting to have 400 users in the future, We would require an aggregate of 4 workers. 


Q5. What are the periods of the performance testing life cycle? 


Before you show up for a performance testing interview, You should know the periods of the performance testing life cycle. 


Ans: Below is the periods of the Performance Testing Life Cycle: 


Prerequisite Gathering: In this stage, We accumulate the non-practical necessity of the application. 


Test Planning: In this stage, we set up the test plan and test procedure record. 


Test Design: In this stage, we plan the test contents and situations. 


Test Executions: In this stage, we execute arranged load tests, stress tests, endurance tests, and so on 


Examination: In this stage, we dissect the performance test results. 


Revealing: In this stage, we set up the performance test reports. 


Q6. What data do you accumulate in the necessity gathering stage? 


Ans: In this stage, we for the most part share a non-utilitarian prerequisite poll with the customer and request the underneath subtleties: 


1. Application Details: Application Name, Type (Mobile/Web/Desktop), Application Accessibility (Intranet/Internet), and so on 


2. Climate Details: Information about Web, App, and DB workers. 


3. Workload Details: No of users, Workflows, Expected exchange each hour (TPH), and so forth 


4. Performance Targets and SLA: Targeted SLA (Service Level Agreement) for exchange reaction time, API reaction time, CPU usage, and Memory use. 


Q7. How might you choose the number of users for a load test? 


Ans: Generally, we get this data from the customer during the necessity gathering period of the performance testing life cycle. 


Anyway now and again, if the customer doesn't have this data, We utilize the accompanying way to deal with choosing the number of users for a load test. 


In the first place, we check if the application is as of now running underway. If indeed, we can get the current creation client check and conduct utilizing APM/Analytics instruments like Appdynamics, Dynatrace, NewRelic, Google Analytics, and so forth 


If the application under test (AUT) is another application and going live interestingly, we can inquire as to whether they have some other comparable application in the creation from which we can extract the information. 


On the off chance that the customer doesn't have some other comparative application, we can follow the gradual model like first test the application with 100 users, then, at that point 200 users, then, at that point 300 users, till the application arrives at its limit. 


Q8. What are the segments of a performance test plan? 


Ans: Below is the segments of the performance test plan: 


1. Performance Test Objectives: For every business cycle, module, or application, the destinations are recorded and characterized here. The destinations originate from the expected workload, change solicitations, or performance necessities. At a significant level, this segment list talk users, number of exchanges, and reaction times for typical and top loads 


2.Test Scope: The extension contains an itemized breakdown of business measures and the load blending of material. This segment additionally diagrams the part or cycles that are out of degree for this performance test, just as any performance testing types that won't be incorporated. 


3. Acknowledgment Criteria: This segment subtleties the undeniable level prerequisites referenced in the Objectives. It characterizes the ordinary and pinnacle loads, the normal exchange, and reaction times for each material part in the extent of the test. The number of users, number of exchanges, or report preparing models each moment/hour/day are recorded. 


4. Test Approach: The biggest part of the report, the Test Approach characterizes the cycle, timing, testing situations, test script creation and approval, and testing area for every performance test type in scope, like Benchmarking Test, Integrated Test, and Stress/Soak Test. The equipment subtleties and correlation of the creation and test conditions are recorded, and all performance testing devices and related checking measures are characterized in this part also. This segment likewise contains the cycle for taking care of imperfections, blunder measurements, and test results documentation. 


5. Test Schedule: For each testing movement, the beginning date, end date, and needed help are recorded in a table. 


6. Passage and Exit Criteria: This part is a rundown of all exercises that should be performed preceding executing the performance test and the measures to be met before considering the performance test total. It additionally records the people or groups answerable for every movement. 


7. Expectations: This rundown any arranged expectations and portrayals alongside those answerable for finishing and conveying them. 


8. Dangers, issues, suspicions, conditions: All performance restrictions, related dangers, and relief endeavors, suppositions, and conditions are depicted in these segments. 


Q9. What is little's law? 


Ans: Little's Law is extremely helpful to characterize the workload model in performance testing. As per the Little's Law: 


N=TPS*(RT+TT+PT) 


N = No of users (Expected virtual users) 


TPS= Transaction each second. (on the off chance that you have TPH-Transaction Per Hour esteem, partition it by 3600 to change over into TPS). 


RT= Total execution season of a script(This incorporate a start to finish execution season of content without think time and pacing) 


TT= Total Think Time in the content (Sum of think time between every one of the exchanges) 


PT= Pacing Time (It is stop time between two cycles, After the finish of the past emphasis and before the beginning of the new cycle) 


Q10. How might you choose the number of load generators needed for a test? 


Ans: One of the most ideal approaches to discover the load generator capacity regarding a test is to run a little test (possibly 5 min) with 5 or 10 users and catch the load generator memory and CPU use subtleties and utilize that data to figure the no of load generator required. 


How? For instance, You ran a test with 10 users for 5 min and saw that LG's memory usage is 200 MB (memory use during test short memory use before beginning the test which is memory saved by OS). It implies your 10 users are devouring 200 MB of your RAM (i.e 20 MB for every client) 


When we have the above data, We can utilize the underneath equation to figure the capacity of one load generator. 


Several users(N) = (75 % of (Total memory of LG - Memory saved by OS) )/Memory used by one client. 


Several load generators = Total no of users/users per load generator.