Performance Analysis of The Bus Topology Network For Effectual Data Distribution
Performance Analysis of The Bus Topology Network For Effectual Data Distribution
International Journal of
Engineering and Applied Computer Science
Volume: 04, Issue: 02, March 2022 ISBN: 9780995707542
10.24032/IJEACS/0402/007
© 2022 by the author(s); licensee Empirical Research Press Ltd. United Kingdom. This is an open access article
distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license.
(http://creativecommons.org/licenses/by/4.0/).
www.ijeacs.com 25
Ojo Jayeola Adaramola et al. International Journal of Engineering and Applied Computer Science (IJEACS)
Volume: 04, Issue: 02, ISBN: 9780995707542, March 2022
subsequent generation of the network. In [10], analytical were HTTP, FTP, email, and databases, which were compared
research on different existing topologies to have a fleeting in the OPNET simulation environment to achieve a reasonable
knowledge of each topology and its characteristics was carried result.
out. In their work, both the advantages and disadvantages of
each topology were presented. They, however, observed that III. METHODOLOGY AND PROCEDURES
two or more of these topologies with their characteristics can
be combined to form a hybrid topology. This topology proved This aspect describes the method and procedure deployed to
to be more effective, reliable, flexible, and scalable as appreciate the performance of well-organized network
advantages, but it required expensive infrastructure and design topology (Bus Topology) for maximum data communication.
complexity. Researchers in [11], also carried out the analytical The process was designed and implemented in an OPNET
investigation on different categories of network topologies that simulation environment where different scenarios were
were based on their advantages, disadvantages, and other modeled. The simulation approach was considered to ease the
differentiating factors that segregate them. A comparison table cost of execution and, thus, demonstrated the expected
was prepared to determine how data flow in the topologies. outcome in the real-life exploitation.
They presented their work as a great tool for dealing with any The office scale (100 x 100 meters) was designed using the
problem related to network topology. In [12], a topology for startup wizard, and the ethcoax model family was selected.
improving performance for their campus, called hybrid Different sizes of bus topologies (10, 20, 30, and 40) were
topology was proposed. This required the comparison of the generated in the rapid configuration of the OPNET, where
output performance of the existing topology with the output ethcoax_station was chosen as the node model, eth_coax as the
performance of the hybrid topology. They concluded that this link model, and eth_tap as the tap model, and other attributes
method produced better and required advancement in network remained in their default position. The bus attributes were
performance for their campus after properly implementing their adjusted with the name changed as eth_coax_adv, the data rate
proposed approach in a simulation environment. of 500,000bps was fixed and the delay of 0.05 seconds was
This was based on the fact that there was no packet loss; considered. Application configuration, profile configuration,
their work was faced with a smaller delay time and had a lower and nodes were configured with (HTTP, FTP, email, and the
hop count. They stressed that this network can be extended to database) to generate traffic. The simulation was set up, and the
handle increasing user numbers. According to the work in [8], global statistics (Delay in (sec), traffic-received in (bits/sec),
performance evaluations on bus topologies using a simulation and traffic-sent in (bits/sec)), and also object statistics
approach were carried out. Their method was based on various (Throughput in (bits/sec), Bits Error per Packet), were
network parameters such as delay in (sec), throughput in obtained.
(bits/sec), traffic-sent in (bits/sec), traffic- received in (bits/sec) Application configuration (default), profile configuration
for 10 Ethcoax stations. An OPNET software was considered (configured with four (4) selected network parameters), and a
the simulation of choice for the implementation and thus server configured to provide required services such as HTTP
simulated for 1 hour. server, FTP server, email server, and database server to ten
They concluded that using the simulation approach is far nodes connected to the eth_coax backbone are shown in Figure
easier than the real-time execution strategy. And they further 1. The model was simulated for 1 hour (60 min) for optimal
stressed that as the number of nodes increases, the network results.
performance decreases and thus experiences little differences in
delay (bits/sec), but there was not much difference in end-to-
end delay. [13], considered three main topologies to include
bus, ring, and star for various numbers of workstations using
software methods. Four different scenarios were examined for
each topology, and each scenario consisted of 5, 10, 15, and 20
workstations. They based their analysis on four different
parameters, which include the number of collisions
experienced on one server, delay in (sec) for global statistics,
load in (bits/sec), and traffic received in (bits/sec) on object
statistics. Each topology was compared distinctly and also
compared for an equal number of linked devices. They came to
the conclusion that as the number of linked workstations
grows, the network's performance declines, and that the bus
topology outperforms the other two topologies. Figure 1. Bus Topology Modelled with 10 Nodes
In this research work, bus topology was investigated with
different sizes of four modeling scenarios: 10 nodes, 20 number Figure 2 shows a server configured to provide the required
nodes, 30 number nodes, and 40 numbers nodes. This method services which are HTTP Server, FTP Server, Email Server,
was used to examine how the bus topology behaved in the and Database Server to the 20 nodes connected to the eth_coax
presence of high packet loss when the number of nodes in the backbone. The model was simulated for 1 hour (60 min) for
network grew. The metrics used to generate network traffic optimal results.
Packet for object statistics. The results were obtained and After one hour of simulation, the results were obtained and
presented in tabular form as shown in Table II after simulating presented in tabular form as shown in Table IV. From this
for one hour. From Table II, the lower delay was observed Table, the higher delay was observed from 20 sec to 30 sec and
between 0-30Secs, more traffic received between 40-60secs, decreased with time, greater traffic was received from 40 sec to
greater traffic sent, and higher throughput experienced from 60 sec, greater traffic was sent at 10 sec, higher throughput was
10-30sec, and error increases as simulation time increase. experienced at 10 sec, and more errors per packet were
observed at 0 sec and decreased on the network as simulation
Table III represents the results obtained for a bus topology time increased.
modeled with 20 nodes. The network parameters of
consideration were delay in (sec), traffic received in (bits/sec), Table V represents the results obtained for a bus topology
and traffic sent in (bits/sec) considering global statistics, modeled with 40 nodes. The network parameters of
throughput in (bits/sec), and Bits Error per Packet for object consideration were delay in (sec), traffic received in (bits/sec),
statistics. After one hour of simulation, the results were traffic sent in (bits/sec) for global statistics, throughput in
obtained and presented in tabular form as shown below. From (bits/sec), and Bits Error for object statistics.
Table III, the higher delay was observed at 0 sec and decreased
with time, greater traffic was received at 10 sec and then, from TABLE V. BUS TOPOLOGY WITH 40 NODES
40–60 sec, higher traffic was sent at 10 sec, higher throughput
was experienced at 110 sec, and errors increased on the Time Delay Traffic Traffic Throughpu Bits
network as simulation time linearly increased. in in Received Sent in t in Error
(sec) (sec) in (bits/sec) (bits/sec) Per
TABLE III. BUS TOPOLOGY WITH 20 NODES (bits/sec) Packet
Time Delay Traffic Traffic Throughpu Bits 0 0.2677 259,640.89 295,367.11 284,666.66 306.28
in in Received in Sent in t in Error 10 0.2635 303,834.57 332,458.67 325,229.83 265.19
(sec) (sec) (bits/sec) (bits/sec) (bits/sec) Per 20 0.2647 303,286.04 331,561.45 324,360.00 269.67
Packet 30 0.2639 303,016.47 331,039.79 323,674.51 272.20
0 0.2748 135,623.11 151,779.56 152,366.67 88.69 40 0.2635 303,492.18 331,665.57 324,158.33 274.79
10 0.2731 152,737.68 166,229.33 166,959.65 116.83 50 0.2641 303,272.66 331,310.18 323,968.24 272.71
20 0.2723 151,961.60 165,114.31 165,653.33 125.71 60 0.2638 302,785.42 330,870.33 323,456.00 273.76
30 0.2722 151,788.48 164,846.50 165,357.05 128.83
40 0.2715 152,161.05 164,954.35 165,416.18 129.86 The results were obtained and presented in tabular form as
shown above after being simulated for one hour. Table V
50 0.2721 152,470.25 165,333.84 165,853.33 128.76 shows that the higher delay was observed at 0 sec and varied
60 0.2721 152,023.04 164,998.26 165,358.67 133.05 time, that more traffic was received at 10 sec, that greater
traffic was sent at 10 sec, that higher throughput was observed
Table IV represents the results obtained for a bus topology at 10 sec and varied time, and that more errors per packet were
consists 30 number nodes. The network parameters of observed and varied time as the simulation time increased.
consideration were delay in (sec), traffic received in (bits/sec), The delay (sec) for each scenario is listed in Table VI, and
and traffic sent in (bits/sec) based on global statistics, the graphical representation of this Table is shown in Figure 6.
throughput in (bits/sec), and Bits Error for object statistics.
TABLE VI. DELAY IN SECS FOR ALL SCENARIOS
TABLE IV. BUS TOPOLOGY WITH 30 NODES
Time in Delay (sec) Delay (sec) in Delay (sec) in Delay (sec) in
Time Delay Traffic Traffic Throughp Bits
(sec ) in for 10 for 20 Nodes for 30 Nodes for 40 Nodes
in in Received in Sent in ut in Error Per
Nodes
(sec) (sec) (bits/sec) (bits/sec) (bits/sec) Packet
0 0.2889 0.2748 0.2650 0.2677
10 0.2872 0.2731 0.2670 0.2635
0 0.2650 198,428.44 222,549.33 512,600.00 260.33
20 0.2886 0.2723 0.2673 0.2647
10 0.2670 228,609.50 249,221.24 246,215.79 218.12
30 0.2884 0.2722 0.2673 0.2639
20 0.2673 227,999,66 247,905.52 245,100.00 208.76
40 0.2912 0.2715 0.2672 0.2635
30 0.2673 228, 540.17 248,678.07 245,547.44 214.35
50 0.2912 0.2721 0.2670 0.2641
40 0.2672 228,813.80 248,684.76 245,720.59 213.27
60 0.2921 0.2721 0.2670 0.2638
50 0.2670 228,805.77 248,656.65 245,661.57 213.08
60 0.2670 228,859.42 248,715.95 245,765.33 213.03
Table IX shows the throughput (bits/sec) for all of the 30 55.61 128.83 214.35 272.20
scenarios, and Figure 9 shows the graphical representation of 40 56.36 129.86 213.27 274.79
this Table.
50 56.00 128.76 213.08 272.71
60 56.07 133.05 213.03 273.76
TABLE IX. THROUGHPUT IN BITS/SEC FOR ALL SCENARIOS
Time Throughput Throughput Throughput Throughput Figure 10 shows that a bus topology with 40 nodes had
in for 10 Nodes for 20 Nodes in for 30 Nodes for 40 Nodes higher bit errors per packet during the data transmission
process than a topology with 10 nodes, which had the lowest
(sec) in (bits/sec) (bits/sec) in (bits/sec) in (bits/sec)
bit errors per packet.
0 77,233.33 152,366.67 512,600.00 284,666.66
10 83,598.15 166,959.65 246,215.79 325,229.83
20 83,413.33 165,653.33 245,100.00 324,360.00
30 83,290.85 165,357.05 245,547.44 323,674.51
40 84,027.45 165,416.18 245,720.59 324,158.33
50 84,140.00 165,853.33 245,661.57 323,968.24
60 84,168.00 165,358.67 245,765.33 323,456.00
V. CONCLUSION
Considering the vital and prevalent concern of distributed
computers with the ability to provide great performance at a
moderate cost, different computers were linked to provide
effective data distribution. However, based on the results
obtained from the simulation environment, it is crystal clear
Figure 9. Throughput in Bits/Sec for all Scenarios that delays were higher in scenarios with lower nodes. But this
result was contrary to the presumption that scenarios with
The Bits Error per Packet for all of the scenarios is fewer nodes would experience little or no delay. Furthermore,
presented in Table X below, with a graphical representation in the scenario with higher nodes sent and received more traffic
Figure 10. than those with lower scenarios. The scenario with a larger
number of nodes experienced higher throughput as a result of
TABLE X. BITS ERROR PER PACKET FOR ALL SCENARIOS the longer backbone (eth_coax) and also encountered the
problem of higher bit errors per packet because of the high
Time Bits Error Bits Error Bits Error Bits Error
collision rate experienced in the network with larger nodes. In
in Per Packet Per Packet Per Packet Per Packet the nearest future, more research into the performance of this
(sec) for 10 Nodes for 20 Nodes for 30 Nodes for 40 Nodes topology could be done by taking into account, a variety of
0 25.30 88.69 260.33 306.28 other network parameters to learn more about the bus topology
performance.
10 38.55 116.83 218.12 265.19
20 50.34 125.71 208.76 269.67