Response Time:
The time it takes between request and complete download of response
How to analyze the scenario for which response has to calculated.
Most of the people will give the priority to performing the performance
of login scenerio.First we should analyze which is the critical functionality
in the application.
For Example :
In Banking application after logging most of the people transfer the
amount . Such kind of Functionalities response time has to be
Performed along with login scenario.
Most of the testers will be worried about what should be maximum
response time of the request. Max is 10 sec.It also depends on
requirement and Business needs.
When total response time exceeds the time given in the requirement. Then the test case is said to be failed.
Developers will check the code through profiler.
Profiler is a program that examines your application as it runs.
It provides you with useful run time information such as time spent
in particular code blocks, memory / heap utilization, number of
instances of particular objects in memory, etc.
When there is no problem with the code ,developer has to check the
architecture or configuration of data base or else they have to tune
the server.
Response time increases exponentially as load increases,and
response time should be less for a request.
Throughput:
It determines the capability of a system or product in
handling multiple request No of multiple request per sec
and Throughput should be less.
If throughput is more than the mentioned in the requirement .
Developer has to tune the application server.
JMeter measures throughput as samples (requests) or
samples or received data per unit time. It can be per min
or per sec.
Latency:
The delay caused by the application ,Operating system and by the environment that are calculated separately.
For example:
For a request
Load Time: 1624
Latency: 1208
Actual response time/load time Should be
Load Time - Latency
1624-1208=416
But due to some problems in the application or operating
system or by the environment a delay is caused that is
known as latency.
Median:
Median is a number which divides the samples into two
equal halves. Half of the samples are smaller than the
median,and half are larger.
For example Sample times for the requests will be
1624
1625
430
1619
430
1651
433
434
454
428
In above example 454 is the median.
Median is used in JMeter just to know server's half response
time is lower than this number.
The requests response time below median value is the
Better response than the request response time above the
median value
Standard Deviation:
Standard Deviation represents how much data varies
from the mean.
In above example median is 454
For first sample request, that is 1624
σ = 1/n * √ Σi=1…n (xi-μ)2
1/1 * √(1624-454)2
=1170(request varies 1170 from the mean)
For the third sample request
σ = 1/n * √ Σi=1…n (xi-μ)2
1/1 * √(430-454)2
=6.92
In the first case, the standard deviation is mostly far from the
mean value and in the second case, the standard deviation
mostly close to the mean value.
The first case is not that much significant than the second
case.
mean value and in the second case, the standard deviation
mostly close to the mean value.
The first case is not that much significant than the second
case.
