• Walang Nahanap Na Mga Resulta

Earth Quake Data Scheduling Using Starvation Free Scheduling Scheme

N/A
N/A
Protected

Academic year: 2023

Share "Earth Quake Data Scheduling Using Starvation Free Scheduling Scheme "

Copied!
24
0
0

Buong text

(1)

Earth Quake Data Scheduling Using Starvation Free Scheduling Scheme

Dr. Priscilla Joy 1, Dr. S. Shirly 2, Dr. R. Venkatesan 3*, Dr. K. Ramalakshmi 4

1 Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Coimbatore.

2 Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Coimbatore.

3* Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Coimbatore.

4 Department of Computer Science and Engineering, Alliance University, Bangalore.

1 rpj.joy@gmail.com, 2 amirja.shirly@gmail.com, 3* rlvenkei2000@gmail.com,

4 ramalakshmivenkatesan@gmail.com

Article Info

Page Number: 1031-1054 Publication Issue:

Vol. 71 No. 3s (2022)

Article History

Article Received: 22 April 2022 Revised: 10 May 2022

Accepted: 15 June 2022 Publication: 19 July 2022

Abstract

The emergence of sensor nodes leads to the need for scheduling of packets in real time. This will lead to efficient packet delivery with less delay and less power consumption. In the existing multilevel scheduling schemes, First Come First Serve Scheduling (FCFS) is commonly used to schedule the real-time packets. FCFS leads to starvation. Hence, Starvation Free Dynamic Multilevel Packet (SF-DMP) scheduling scheme is applied. Using this an earthquake data is scheduled which is taken from the dataset. Furthermore, the performance is high due to the emergent data from the earthquake dataset will be send fast to any system assigned there by an alert can be sent.

Keywords: Scheduling, Priority, Earth quake, Scheme, Level, Starvation.

1. Introduction

It is critical to use packet scheduling at sensor nodes synonymously with task scheduling in order to deliver various types of data packets based on their priority and impartiality with the least amount of latency, among other network design problems, such as routing protocols and data aggregation, in order to reduce sensor energy consumption and data transmission delay.

Sensitive data, such as that for real-time applications, is given precedence over less critical data. Many studies on sensor node sleep-wake time scheduling have been done; however, the literature on sensor node packet scheduling, which schedules processing of available data packets and also cuts energy use, is scant.

According to current Wireless Sensor Network (WSN) operating systems, data packets are processed in the order in which they arrive, which means that they take a long time to arrive at a base station (BS). Data must arrive to the BS in a specified time frame or before a deadline to be considered significant. In addition, the minimum end-to-end latency feasible should be used to provide real-time emergency data to BS. As a result, intermediary nodes need to reorder their ready queue based on the relevance of the data packets in their queue (e.g., real or non-real time).

(2)

Due to their present and static nature, most existing WSN packet scheduling algorithms aren't dynamic or appropriate for large-scale applications. They also can't be altered in response to changes in application needs or conditions. For example, a real-time priority scheduler is statically utilised in many real-time applications and cannot be altered while WSN apps are running.

2. Existing Work

In the existing Dynamic Multilevel Packet (DMP) Nasser et al., [2013] scheduled using FCFS. The amount of real time packets arriving is high in priority. As the previous scheme uses FCFS for the same purpose, it suffers from starvation due to the real time packets waiting more which is emergent may arrive later than the other packets. This leads to starvation. Previously, the priority is assigned based on hop count of the nodes as of the base station. So, the nodes with the highest hop count are given the preference. Due to this assignment, the emergent packet which is located in the nearest hop count will be waiting for a long time to be scheduled which will lead to starvation. To remedy this problem the algorithm was introduced.

Lipika [2015] introduces a novel round robin scheduling method for real time. Nayak et al., [2012] and Negi introduced the dynamic time quantum mechanism. Rajput and Gupta [2012]

introduce a priority based round robin scheme. Varma [2013] suggest relative performance analysis is considered and found.

Iraji et al., [2015] suggested the dynamic weighted harmonic round robin. Behera et al., [2011]

suggested re-adjusted round robin scheduling algorithm. Mostafa et al., [2010] given a new time quantum measurement for the round robin mechanism. In Mohanty et al., [2011]; Leung and Merrill [1980] the performance evaluation is suggested. Baital and Chakrabarti [2016]

suggest about the preemptive scheduling. In Kinsy and Devadas [2014], Salamy et al., [2013]

and Khurma et al., [2018] dynamic scheduling in tasks were done.

Datta [2015] gives an efficient algorithm through dynamic time slice. In Saeidi and Baktash [2012], Farooq et al., [2017] the time quantum based scheduling is carried out. In Chochiang et al., [2019] Weighted Priority Scheduling is done. Dave et al., [2017] suggests the modified round robin scheme. In Ayeni et al., [2017], Derahman et al.,[2016], Elmougy et al., [2017]

and Fataniya and Patel [2018] dynamic time quantum is used for carrying out the round robin scheduling.

Marosits et al., [2001] suggests about the round robin scheme and based on which the quality of service is maintained. Yuan and Duan [2005] suggest a round robin scheduler. Arif et al., [2016] implemented an alternating median based round robin scheduling algorithm.

Chahar and Raheja [2013] suggested that the ready queue is categorized into two queues based on the CPU bound and input output bound. By scheduling based on the above categories the starvation which occurred previously has removed.

Madhumathi and Kalaiyarasi [2018] suggest a remedy for the resource allocation issue. They usedx the Shortest Job First (SJF) scheduling. This leads to the starvation. This shortcoming is reduced by combining both the SJF and multilevel queue scheduling.

Thombare et al., [2016] suggest that the multilevel feedback queue for small task is analogous to round robin scheduling. By implementing the concept of dynamic time quantum in round robin, the starvation occurred has removed. It will further improve the performance.

(3)

Biswas et al., [2017] suggest that multilevel queue scheduling s proving the parallelizing of the subtasks. By implementing the resource utilization is shared fairly.

K and Gupta [2014] suggest that by using the Multilevel Feedback Queue (MFQ) each and every ready queue used different scheduling techniques. As a result, the performance is high compared to the previous schemes.

Scheduling is required for arranging the nodes and to send the packets earlier. By introducing the scheduling, the power consumption issue which arises will be reduced. Many researchers have proposed algorithms for scheduling in wireless sensor networks. I have referenced a lot of papers in scheduling and I have quoted some of it here.

3. Proposed System

In the scheduling scheme the overall architecture comprises of the four steps as in figure 3.1:

• Zone Classification

• Node Level Classification

• Priority Level Classification

• Scheduling Phase

Figure 3.1; Overall Architecture

The nodes are categorized as node levels based on the position in the zones as in Figure 3.2. In SF-DMP scheduling scheme depicted in figure 3.2 the packets were first classified as four zones and in each and every zone the task was identified in the task identification zone.

(4)

Figure 3.2; SF-DMP Scheduling Scheme

Then it is fed into the queues in which the tasks were using minheap or maxheap technique and based on the order it will choose either maxheap or minheap and is placed in the queue.

After that in the scheduling phase based on the priority levels the scheduling is done. The sum of the burst times of task is equal to the time quantum is calculated. The modulo value is generated as burst time modulo time slice. Founded on this scheme the priority level 1 packets which contains the real time urgent task were scheduled using round robin scheduling.

The Pr2 packets which contains the real time task were scheduled using round robin scheduling. Pr3 task is scheduled using round robin scheduling.

Algorithm 1 SF-DMP Algorithm 1: Procedure SF-DMP ALGORITHM Do

begin

while task arrives

if Type_of_task = real_time then place task in Pr1 queue

else if Type_of_task = non_real_time and remote then place task in Pr2 queue else

place the task in Pr3 queue end if

if(maxheap==1)

if (right ≤ N and Arr[right] >Arr[longest]) if (left ≤ n and Arr[left] >Arr[i]) Longest = left;

else

Longest = j Longest = right; if (longest! =j)

Swap (Arr[i], Arr[longest]); Max_heapify (Arr, longest, N);

for (int j= N/2); j ≥1; j–);

Max_heapify(Arr,j);

if(minheap==1)

(5)

if (left ≥ n and Arr[left] <Arr[j]) Small= left;

else Small =j

if (right ≥ N and Arr[right] <Arr[small]) small = left;

if (small! =j)

Swap (Arr[j], Arr[small]);

1: end procedure

2: procedure SF-DMP ALGORITHM Min_heapify (Arr, small, N);

for (int j= N/2); j ≥1; j–);

Min_heapify (Arr,j); for Pr1 task:

I/P : Process (Pn), Burst Time (BTi), Highest Burst Time (HBT), priority(Pi), Ready Queue(RQ) ;

O/P: Context Switch (CS), Average Waiting Time(Awt), Average Turnaround Time(Atat).

Initialize Ready Queue,CS,Awt, Atat, Quantum Time (QT) as 0 Let n=number of processes;

for (j=1to n) Priority ratio (PRj)= 0; Burst ratio (BRj)= 0 Remaning Burst Time (RBTj)= 0;

Precedence Factor (PFj)=0; Median=0;

for (j=1 to n) RBTj = BTj

for (j=1 to n) PFj = (Pj x PRj) + (BTj x BRj) If n is odd then:

median= Total number o f processes If n is even then:

median= Total number o f processes+1 QT=ceil (sqrt(median*HBT)) Modulo Burst Time=

BTj%QT End while

2: end procedure

The flowchart of SF-DMP scheduling scheme is given in Figure 3.4.

(6)

Figure 3.4; Flowchart of SF-DMP Scheduling Scheme

The C programming language is used for simulation. From USGS science for a changing world which contains the spreadsheet format about the earth quake real time notifications feeds and services, the parameters like magnitude and time were taken from the dataset.

From USGS science for a changing world which contains the spreadsheet format about the earth quake real time notifications feeds and services, the parameters like magnitude and time were taken from the dataset. It was observed that if the magnitude increases more than 2.5 MW, it will lead to earthquake. So, the values more than 2.5 mw is taken as real time data (Pr1) in CW-DMP, ITF, IPS and SF-DMP scheduling scheme. For pr2 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr3, the value which is below 1mw is taken.

But in EDMP the pr1 contains real-time urgent data. So, the values above 3mw which causes major disaster will be considered as real-time urgent data and placed in Pr1 queue. Pr2 contains the real time data which is ranging between 2.5 mw and 3 mw. For pr3 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr4, the value which is below 1mw is taken.

Table 4.1. Simulation Parameters and their Values Parameter Values

Network Size-100m x 100m Number of nodes (Maximum 200) Number of zones-4

Number of tasks-10000

Base station position-55m x 101m

(7)

Transmission energy consumptions-50 nJoule/bit

Energy consumption in free space or Air -0.01 nJoule/bit/m2 Initial node energy - 2 Joule

Transmission speed - 250 Kbps

Propagation speed -198 x 106 meter/sec Task priority levels -3

Time Quantum – 5 Zone height - 20 4. Results and Discussion

The C programming language is used for simulation. From USGS science for a changing world which contains the spreadsheet format about the earth quake real time notifications feeds and services, the parameters like magnitude and time were taken from the dataset.

From USGS science for a changing world which contains the spreadsheet format about the earth quake real time notifications feeds and services, the parameters like magnitude and time were taken from the dataset. It was observed that if the magnitude increases more than 2.5 MW, it will lead to earthquake. So, the values more than 2.5 mw is taken as real time data (Pr1) in CW-DMP, ITF, IPS and SF-DMP scheduling scheme. For pr2 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr3, the value which is below 1mw is taken.

But in EDMP the pr1 contains real-time urgent data. So, the values above 3mw which causes major disaster will be considered as real-time urgent data and placed in Pr1 queue. Pr2 contains the real time data which is ranging between 2.5 mw and 3 mw. For pr3 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr4, the value which is below 1mw is taken.

Table 4.1. Simulation Parameters and their Values Parameter Values

Network Size-100m x 100m Number of nodes (Maximum 200) Number of zones-4

Number of tasks-10000

Base station position-55m x 101m

Transmission energy consumptions-50 nJoule/bit

Energy consumption in free space or Air -0.01 nJoule/bit/m2 Initial node energy - 2 Joule

Transmission speed - 250 Kbps

Propagation speed -198 x 106 meter/sec Task priority levels -3

Time Quantum – 5 Zone height - 20

(8)

4.1. Case Study of DMP

In DMP the pr1 contains real-time urgent data. So the values above 3mw which causes major disaster will be considered as real-time urgent data and placed in Pr1 queue. Pr2 contains the real time data which is ranging between 2.5 mw and 3 mw. For pr3 queue, the magnitude ranging below 2.5mw will be considered. The snapshot of Pr1,Pr2 and Pr3 are shown in Figure 4.1, Figure 4.2 and Figure 4.3 respectively.

Figure 4.1; Pr1 Snapshot of DMP

Figure 4.2; Pr2 Snapshot of DMP

(9)

Figure 4.3; Pr3 Snapshot of DMP

Figure 4.4; Case Study of DMP

(10)

Table 4.2. Case Study of DMP

Case study of DMP Scheduling scheme

Priority Level Average Waiting Time (ms) Average Turnaround Time (ms)

Pr1 26.73 26.73

Pr2 90.29 90.29

Pr3 26.35 26.35

The case study of DMP is tabulated in Table 4.2 and the graph is plotted in Figure 4.4.

4.2. Case Study of EDMP

Figure 4.5; Pr1 Snapshot of EDMP

In EDMP the pr1 contains real-time urgent data. So, the values above 3mw which causes major disaster will be considered as real-time urgent data and placed in Pr1 queue. Pr2 contains the real time data which is ranging between 2.5 mw and 3 mw. For pr3 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr4, the value which is below 1mw is taken. The snapshot of Pr1, Pr2, Pr3 and Pr4 are shown in Figure 4.5, Figure 4.6, Figure 4.7 and Figure 4.8 respectively.

Figure 4.6; Pr2 Snapshot of EDMP

(11)

Figure 4.7; Pr3 Snapshot of EDMP

Figure 4.8; Pr4 Snapshot of EDMP

(12)

The case study of EDMP is tabulated in Table 3.3.

Table 4.3; Case study of EDMP Case study of EDMP Scheduling scheme

Priority Level Average Waiting Time (ms) Average Turnaround Time (ms)

Pr1 8.32 12.64

Pr2 1.29 3.89

Pr3 90.29 90.29

Pr4 26.35 26.35

Based on the tabulation above, the graph for the average waiting time and average turnaround time for EDMP is plotted in Figure 4.9.

From the table above the pr1 and pr2 packets were take the mean and considered as the real time packets. Thus it is compared as three priority queues as like the other scheduling schemes and compared accordingly.

Figure 4.9: Case Study of EDMP

4.3. Case Study of CW-DMP

In CW-DMP, the values more than 2.5 mw is taken as real time data (Pr1). For pr2 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr3, the value which is below 1mw is taken. Based on the magnitude the waiting time and turnaround time for the three priority levels were computed and the snapshot is given in Figure 4.10, 4.11 and 4.12 respectively.

Figure 4.10; Pr1 Snapshot of CW-DMP

(13)

Figure 4.11; Pr2 Snapshot of CW-DMP

Figure 4.12; Pr3 Snapshot of CW-DMP

Figure 4.13; Case Study of CW-DMP

(14)

The case study of CW-DMP is tabulated in Table 3.4 and the graph is plotted in Figure 4.13.

Table 4.4. Case Study of CW-DMP Case study of CW-DMP Scheduling scheme

Priority Level Average Waiting Time (ms) Average Turnaround Time (ms)

Pr1 9.22 13.03

Pr2 34.9 36.47

Pr3 0 0

4.4. Case Study of ITF

In ITF, the values more than 2.5 mw is taken as real time data (Pr1). For pr2 queue,

Figure 4.14; Pr1 Snapshot of ITF

The magnitude ranging between 1mw and 2.5mw will be considered. In Pr3, the value which is below 1mw is taken.

Figure 4.15; Pr2 Snapshot of ITF

(15)

Figure 4.16; Pr3 Snapshot of ITF

Based on the above Figure 4.14, Figure 4.15 and Figure 4.16 the following Table 4.5 is tabulated.

Table 4.5; Case Study of ITF Case study of ITF Scheduling scheme

Priority Level Average Waiting Time (ms) Average Turnaround Time (ms)

Pr1 9.22 13.03

Pr2 38.02 39.59

Pr3 26.35 26.35

Based on the values calculated from the above table 5.36 the graph is plotted in the Figure 4.17.

(16)

Figure 4.17; Case Study of ITF

4.5. Case Study of IPS

In IPS, the values more than 2.5 mw is taken as real time data (Pr1). For pr2 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr3, the value which is below 1mw is taken. The snapshot of Pr1, Pr2 and Pr3 is given in figure 4.18, figure 4.19 and figure 4.20 respectively.

Figure 4.18; Pr1 Snapshot of IPS

(17)

Figure 4.19; Pr2 Snapshot of IPS

Figure 4.20; Pr3 Snapshot of IPS

4.6. Case Study of SF-DMP

In SF-DMP, the values more than 2.5 mw is taken as real time data (Pr1). For pr2 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr3, the value which is

(18)

below 1mw is taken. The snapshot of Pr1, Pr2 and Pr3 is given in Figure 4. 22, Figure 4. 23 and Figure 4.24 respectively.

Figure 4.22; Pr1 Snapshot of SF-DMP

Figure 4.23; Pr2 Snapshot of SF-DMP

(19)

Figure 4.24; Pr3 Snapshot of SF-DMP

Based on the data from the above snapshot the following Table 4.7 is tabulated and the graph is plotted in Figure 4.25.

Table 4.7. Case Study of SF-DMP Case study of ITF Scheduling Scheme

Priority Level Average Waiting Time (ms) Average Turnaround Time (ms)

Pr1 7.89 11.72

Pr2 31.3 32.86

Pr3 0 0

As the burst time ranges 0 to 1ms, the waiting time and there by the turnaround time also falls the same. According to different applications the value varies. But compared to all other schemes - the value diminishes in SF-DMP.

(20)

Figure 4.25; Case Study of SF-DMP

Figure 4.26; Comparison of Scheduling Schemes Based on the Case Study

Based on the case study from the real time earth quake data the average waiting time and turnaround time were calculated as follows in Table 4.8 and the graph is plotted in Figure 4.26.

CW-DMP and SF-DMP values are nearby due to the use of round robin scheduling and the value reached 0 ms in both the cases.

Table 4.8. Average Waiting Time and Turnaround Time of Earthquake Real Time Data Scheduling Schemes Average Waiting Time (ms) Average Turnaround Time (ms)

DMP 47.79 47.79

EDMP 31.57 33.29

CW-DMP 14.46 16.49

ITF 24.53 26.32

IPS 21.53 23.57

(21)

SF-DMP 12.82 14.86

4.7. Case Study of SF-DMP

In SF-DMP, the values more than 2.5 mw is taken as real time data (Pr1). For pr2 queue, the magnitude ranging between 1mw and 2.5mw will be considered. In Pr3, the value which is below 1mw is taken. Based on the data from the above Table 4.9 is tabulated and the graph is plotted in Figure 4.27.

Table 4.9. Case Study of SF-DMP

Case study of SF-DMP Scheduling Scheme

Priority Level Average Waiting Time (ms) Average Turnaround Time (ms)

Pr1 7.89 11.72

Pr2 31.3 32.86

Pr3 0 0

Figure 4.27; Case Study of SF-DMP

As the burst time ranges 0 to 1ms, the waiting time and there by the turnaround time also falls the same. According to different applications the value varies. But compared to all other schemes - the value diminishes in SF-DMP.

Figure 4.28; Comparison of Scheduling Schemes Based on the Case Study

(22)

Based on the case study from the real time earth quake data the average waiting time and turnaround time were calculated as follows in Table 4.9 and the graph is plotted in Figure 4.28.

CW-DMP and SF-DMP values are nearby due to the use of round robin scheduling and the value reached 0 ms in both the cases. The below table shows the graphical representation of the case Study of SF-DMP.

Table 4.10; Average Waiting Time and Turnaround Time of Earthquake Real Time Data Scheduling Schemes Average Waiting Time (ms) Average Turnaround Time (ms)

DMP 47.79 47.79

EDMP 31.57 33.29

CW-DMP 14.46 16.49

ITF 24.53 26.32

IPS 21.53 23.57

SF-DMP 12.82 14.86

5. Conclusions

When compared with the previous scheme based on the case study of the earthquake data, the waiting time of EDMP decreases by 16.22ms, CWDMP decreases by 33.23ms, ITF decreases by 23.26ms and IPS decreases by 26.26ms, SF-DMP the waiting time decreases by 37.97ms.

Thus, the starvation is mitigated in the proposed scheme. SF-DMP outperforms by reducing its average waiting time by 5.2ms than IPS packet scheduling scheme. In the comparison made on non-real time of the average waiting time among the scheduling schemes, SF-DMP outperforms by reducing its average waiting time by 46.8ms than DMP and EDMP, 11.7ms than CW-DMP, 6.86ms than ITF and IPS packet scheduling scheme. When compared with the previous scheme based on the case study of the earthquake data, the waiting time of EDMP decreases by 16.22ms, CWDMP decreases by 33.23ms, ITF decreases by 23.26ms and IPS decreases by 26.26ms. Based on the case study, the average waiting time and turnaround time of SF-DMP reduced than all the other proposed scheduling schemes. Thus, it outperforms among the proposed scheduling schemes.

References

1. Nasser N., Karim L. and Taleb T., (2013), Dynamic multilevel priority packet scheduling scheme for wireless sensor network, IEEE transactions on wireless communications, 12(4), 1448–1459.

2. David, H.E., Ramalakshmi, K., Venkatesan, R., Hemalatha, G. (2021) Tomato leaf disease detection using hybrid CNN-RNN model Advances in Parallel Computing, (38), 593-597.

3. Lipika D., (2015), A New RR Scheduling Approach for Real Time Systems using Fuzzy Logic, International Journal of Computer Applications, 119(5).

4. Nayak D., Malla S. K. and Debadarshini D., (2012), Improved round robin scheduling using dynamic time quantum, International Journal of Computer Applications, 38(5), 34–38.

5. Rajput I.S. and Gupta D., (2012). A priority based round robin CPU scheduling algorithm for real time systems, International Journal of Innovations in Engineering and Technology, 1(3), 1–11.

(23)

6. Mohanty R., Das M., Prasanna M.L. et al., (2011), Design and performance evaluation of a new proposed fittest job first dynamic round robin (FJFDRR) scheduling algorithm, arXiv preprint arXiv:1109.3075, 2(2), 23–27.

7. Baital K. and Chakrabarti A. (2016). An efficient dynamic scheduling of tasks for multicore real-time systems. In Advances in computing applications, 31–47. Springer.

8. Datta L., (2015). Efficient round robin scheduling algorithm with dynamic time slice, International Journal of Education and Management Engineering, 2, 10–19.

9. Saeidi S. and Baktash H.A. (2012). Determining the optimum time quantum value in round robin process scheduling method, IJ Information Technology and Computer Science, 10, 67–73.

10. Papagianni, C.; Leivadeas, A.; Papavassiliou, S.; Maglaris, V.; Cervello-Pastor, C.; Monje, A.

On the optimal allo-cation of virtual resources in cloud computing networks. IEEE Trans.

Comput. 2013, 62, 1060–1071.

11. Venkatesan, R., Anni Princy, B., Ambeth Kumar, V.D., Kumar, A., Khan, A.K.(2021) Secure online payment through facial recognition and proxy detection with the help of TripleDES encryption Journal of Discrete Mathematical Sciences and Cryptography, 24(8) 2195-2205.

12. Fataniya B. and Patel M., (2018), Dynamic time quantum approach to improve round robin scheduling algorithm in cloud environment, International Journal of Scientific Research in Science, Engineering and Technology, 4(4), 963–969.

13. Kaewpuang, R.; Niyato, D.; Wang, P.; Hossain, E. A Framework for Cooperative Resource Management in Mobile Cloud Computing. IEEE J. Sel. Areas Commun. 2013, 31, 2685–2700.

14. Xiao, Z.; Song, W.; Chen, Q. Dynamic resource allocation using virtual machines for cloud computing environ-ment. IEEE Trans. Parallel Distrib. Syst. 2012, 24, 1107–1117.

15. Jebaseeli, T.J., David, D.J., Venkatesan, R. (2021) Prediction of COVID'19 through multiple organ analysis using iot devices and machine learning techniques, International Journal of Engineering Trends and Technology 69(8)- 102-108.

16. Warneke, D.; Kao, O. Exploiting Dynamic Resource Allocation for Efficient Parallel Data Processing in the Cloud. IEEE Trans. Parallel Distrib. Syst. 2011, 22, 985–997.

17. Madhumathi R. and Kalaiyarasi N. (2018), Credit Based Multilevel Queue Scheduling in Cloud Environment. In 2018 International Conference on Communication, Computing and Internet of Things (IC3IoT), 87–90. IEEE.

18. Goar, . V. K. ., and N. S. . Yadav. “Business Decision Making by Big Data Analytics”.

International Journal on Recent and Innovation Trends in Computing and Communication, vol. 10, no. 5, May 2022, pp. 22-35, doi:10.17762/ijritcc.v10i5.5550.

19. Son, S.; Jung, G.; Jun, S.C. An SLA-based cloud computing that facilitates resource allocation in the distributed data centers of a cloud provider. J. Supercomput. 2013, 64, 606–637.

20. Wei, G.; Vasilakos, A.V.; Zheng, Y.; Xiong, N. A game-theoretic method of fair resource allocation for cloud com-puting services. J. Supercomput. 2010, 54, 252–269.

21. Laili, Y.; Tao, F.; Zhang, L.; Sarker, B.R. A study of optimal allocation of computing resources in cloud manufacturing systems. Int. J. Adv. Manuf. Technol. 2012, 63, 671–690.

(24)

22. M.Syed Shahul Hameed, N.Suganthi, S.Balakrishnan, Cell-Leach Based Wireless Sensor Network For Optimized Energy Consumption, Turkish Journal of Physiotherapy and Rehabilitation, Vol. 32, Issue 2, pp. 2452-2456.

23. S. Balakrishnan, B. Persis Urbana Ivy and S. Sudhakar Ilango, “A Novel And Secured Intrusion Detection System For Wireless Sensor Networks Using Identity Based Online/Offline Signature”, ARPN Journal of Engineering and Applied Sciences. November 2018, Vol. 13 No. 21, pp. 8544-8547.

24. S.Balakrishnan, Vinod K, B. Shaji. (2018). “Secured and Energy Efficient AODV Routing Protocol For Wireless Sensor Network”, International Journal of Pure and Applied Mathematics, Vol. 119, No. 10c, 2018, pp. 563-570.

25. S.Balakrishnan, J.P.Ananth, L.Ramanathan, S.P.Premnath, (2018). “An Adaptive Energy Efficient Data Gathering In Wireless Sensor Networks”, International Journal of Pure and Applied Mathematics, Volume 118 No. 21, 2018, pp. 2501-2510.

26. J.P.Ananth, S.Balakrishnan, S.P.Premnath, (2018). “Logo Based Pattern Matching Algorithm for Intrusion Detection System in Wireless Sensor Network”, International Journal of Pure and Applied Mathematics, Volume 119, No. 12, 2018, pp. 753-762.

27. M. Syed Shahul Hameed , Ravi Kumar Poluru, Chandra Prakash Lora, S. Balakrishnan, A Non-Invasive Check Alarm Towards Food Safety Against Scruples Using Iot For The Produce, Nat. Volatiles & Essent. Oils, 2021; 8(4): 10734-10743 10734.

28. Philip, A. M., and D. S. . Hemalatha. “Identifying Arrhythmias Based on ECG Classification Using Enhanced-PCA and Enhanced-SVM Methods”. International Journal on Recent and Innovation Trends in Computing and Communication, vol. 10, no. 5, May 2022, pp. 01-12, doi:10.17762/ijritcc.v10i5.5542.

Mga Sanggunian

NAUUGNAY NA DOKUMENTO

The themes were directly derived from the research problems of the study: 1 Performance achievement of the students in the pretest and posttest before and after using game- based