Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

For some projects, testing with simulated data or live data is just a matter of

ID: 3920217 • Letter: F

Question

For some projects, testing with simulated data or live data is just a matter of changing data set files and setting up some new systems configurations. However, for some other projects, selecting either simulated data or live data for testing is a big concern for systems analysts because their decisions would affect other groups. If you are preparing a detailed test plan as a systems analyst, please explain your proposed testing strategies plan regarding the use of either simulated data or live data. Will you use simulated data or live data? Why?

Explanation / Answer

System testing will be definetly more effective if we get the actual data or the live
data because this will tell the actual behaviour of the system. But there may be
many constraints with the live data.Many be the actual system is at different place
or timings are big constraing or integrating the test system in the actual setup
is a problem, so there can be n number of issues with the real data.However simulated
data is easy to manage and generate.There is lot of flexibility with the simulated data
as we can control its generation.Coming to testing , it really depends on what exactly
we are testing and what are the constraints with either live data or simulated data.The
approach should be get as close as to the actual environment.It is always good to start
with the simulated data to establish the initial stability of the system in terms of testing.Once a certain level of testing requirements are fullfilled (more focussing on the accuarcy part rather than performance or other non functional requirements), we
can move the testing of the system slowly in a phased mannere to live data or actual
data based on its availability and constraints.If we don't move in this controlled fashion it will be very difficult to asses the failure situations because in actual
scenario things are not at control and we can not pin point actual reason of the failure.So we should better perform initial testing with controlled set of daat in order to establish clear decisions on system functionality as far as testing is concernes and then to the actual environment or live data to establish the credibility of the system in actual sense.