Information Technology Reference
In-Depth Information
The chart above clearly shows that the response time is at the minimum for
scenario 2 while it is at the maximum in case of scenario 1.
Following figure (Fig 5) is a chart with Number of users plotted on X-axis vs.
Throughput plotted on Y-axis.
From the chart shown in Figure 5 it is clear that scenario 2 throughput > scenario 3
throughput is the best suitable topology:
scenario 2 throughput > scenario 3 throughput
Thus, keeping Apache web server and application server on the same instance and
auto-scaling this server instance is capable of handling more load than the topology
where web server and app server are hosted on separate instances.
Table 2.
Comparison Chart for 3 Scenarios
Avg.
Through-
put
Scenario
No.
Number
of threads
Configura-
bility
Fault
tolerance
Maintaina-
bility
Throughput
Availability
Scalability
25
214.8
50
209.7
No failover
capability
Not
Scalable
1
218.40
Easy
No
Easy
75
211
100
238.1
25
679.1
50
633.5
Highly
Available
2
724.25
Moderate
Yes
Moderate
Scalable
75
731.9
100
852.5
25
313.7
50
437.2
Highly
Available
3
421.32
Difficult
Yes
Difficult
Scalable
75
482.4
100
452
3.5
Observations
•
Scenario 1
: It gives moderate performance as compared to other 2 scenarios. All
3 servers are installed on same instance causing the degradation in performance
and hence the response time. Though the configuration and maintenance of this
setup is very easy, it is devoid of any failover capability. At any given point in
time there would only be one sever which would be serving all the requests to the
application and as the number of requests increases the performance of an
application would go on decreasing.
