Dynamic Resource Allocation Using Improved Firefly Optimization Algorithm in Cloud Environment

Figures & data

Figure 1. DSA-based strategies.

Table 1. Summary of related works.

Reference	Objective	Algorithm used
Skarlat et al. (2016)	A conceptual framework for providing resources in fog computing	A meta-heuristic method based on genetic algorithm
Alelaiwi (2017)	A scalable model based on linear programming in the cloud federation	A multi-objective optimization model based on genetic algorithm
Elhoseny et al. (2018)	A scheduling and load balancing model based on the concept of cloud segmentation	Round robin and game theory
Kumar and Kumar (2019)	An overview of common load balancing algorithms in cloud computing	One of the load balancing algorithms is Min-Min.
Aghdashi and Mirtaheri (2019)	A hybrid scheduling strategy for managing and processing medical data in the cloud	Combining genetic algorithms and particle swarm optimization
Rajabion et al. (2019)	A DSA approach in cloud computing and the Internet of Things for real-time and batch processing	Analysis of different algorithms
Taher et al. (2019)	An autonomous approach based on fuzzy hierarchical technique for DSA in the cloud	Support vector machines and linear regression
Alboaneen et al. (2020)	Joint task scheduling and VM placement	Glowworm swarm optimization, moth-flame glowworm swarm optimization, and genetic algorithm
Aburukba et al. (2020)	A resource scheduling algorithm for managing IoT services using integer linear programming	Improved genetic algorithm
Faraji Mehmandar, Jabbehdari, and Haj Seyyed Javadi (2020)	A distributed computing framework for resource management in fog computing	Reinforcement learning and support vector regression techniques
Wang, Liu, and Jolfaei (2020)	A multi-objective optimization approach for DSA on cloud-based sensor networks	Chaotic firefly algorithm
Mahini et al. (2021)	A scheduling algorithm for multilayer DSA in the cloud	A formal definition of cloud federation integrity based on clustering
Jula et al. (2021)	An approach based on workload analysis to ensure resource efficiency	Imperialist competitive algorithm and fuzzy c-means clustering

Figure 2. Flowchart of FA steps.

Figure 3. IFA-DSA flowchart.

Figure 4. Trapezoidal membership function with three modes to prioritize tasks.

Figure 5. Encoding structure of fireflies.

Table

Algorithm 1. Proposed heuristic method for creating an initial population
	Input: $N_P$: Population size; $M$: Number of VMs; $N$: Number of tasks.
	Output: $Pop\left(N_P,N\right)$ as the initial population.
1:	$Pop\left(N_P,N\right)\leftarrow \mathrm{n}\mathrm{u}\mathrm{l}\mathrm{l}$.
2:	for		$i$ = 1 to $N_P$ do
3:			for		$j$ = 1 to $N$ do
4:					if			Task($t_j$) is high priority then
5:								$t_h\to v_d|min\left({\sum }^{t_h}\right)\in VM$
6:								$Pop\left(i,j\right)=v_d$
7:					elseif			Task($t_j$) is middle priority then
8:								$t_m\to v_d|min\left({\sum }^{t_h}+{\sum }^{t_m}\right)\in VM$
9:								$Pop\left(i,j\right)=v_d$
10:					else
11:								$t_l\to v_d|min\left({\sum }^T\right)\in VM$
12:								$Pop\left(i,j\right)=v_d$
13:					end
14:			end
15:	end
16:	The initial population created is $Pop$, as output.

Table 2. Scenarios defined for the DSA problem.

Scenario	Number of tasks ($N$)	Number of VMs ($M$)
Scenario 1	100	5
Scenario 2	50	10
Scenario 3	500	60
Scenario 4	1000	100

Table 3. Configuring the cloud environment for the DSA problem.

Parameter	Value
Task size	Randomly between 500 and 2000
Power VMs	Randomly between 100 and 1000
Number of resources in VMs	Randomly between 1 and 4

Figure 6. VMs load balancing at the beginning of the scheduling process.

Figure 7. Distribution of tasks with different priorities on VMs.

Figure 8. Comparison of FA with GA and PSO in the proposed method.

Figure 9. Comparison of IFA-DSA with similar methods in the makespan criteria.

Table 4. Statistical analysis of the proposed method in comparison with other methods.

Evaluation criteria	FCFS					TSDQ-FLPSO					ICFA
	Mean	SD	$t$-value	$p$-value		Mean	SD	$t$-value	$p$-value		Mean	SD	$t$-value	$p$-value
Workload balance	2.16	3.73	2.54	0.016		2.95	3.92	5.98	0.009		3.01	2.67	2.29	0.01489058
Makespan	50.16	5.43	2.73	0.018		53.04	5.30	2.64	0.015		53.45	5.23	2.32	0.0256673

Table 5. Parameters used by different methods for DSA.

Parameters	FCFS	TSDQ-FLPSO	ICFA	IFA-DSA
Workload balance	√	√	√	√
Being fair	-	√	-	-
Efficiency	√	-	-	-
Utilization of resources	-	-	√	-
Run-time	-	-	-	√
Energy consumption	-	-	√	-
Processing cost	√	-	-	√
Response time	√	√	√	-
Prioritize tasks	-	-	-	√

Skarlat, O., S. Schulte, M. Borkowski, and P. Leitner (2016, November). Resource provisioning for IoT services in the fog. In 2016 IEEE 9th international conference on service-oriented computing and applications (SOCA), Macau, China, (pp. 32–39). IEEE.

 Google Scholar

Alelaiwi, A. 2017. A collaborative resource management for big IoT data processing in Cloud. Cluster Computing 20 (2):1791–99. doi:10.1007/s10586-017-0839-y.

 Web of Science ®Google Scholar

Elhoseny, M., A. Abdelaziz, A. S. Salama, A. M. Riad, K. Muhammad, and A. K. Sangaiah. 2018. A hybrid model of internet of things and cloud computing to manage big data in health services applications. Future Generation Computer Systems 86:1383–94. doi:10.1016/j.future.2018.03.005.

 Web of Science ®Google Scholar

Kumar, P., and R. Kumar. 2019. Issues and challenges of load balancing techniques in cloud computing: A survey. ACM Computing Surveys (CSUR) 51 (6):1–35. doi:10.1145/3281010.

 Web of Science ®Google Scholar

Aghdashi, A., and S. L. Mirtaheri. 2019, April. A survey on load balancing in cloud systems for big data applications. In International congress on high-performance computing and big data analysis, Lucio Grandinetti, Seyedeh Leili Mirtaheri, Reza Shahbazian editors, 156–73. Cham: Springer.

 Google Scholar

Rajabion, L., A. A. Shaltooki, M. Taghikhah, A. Ghasemi, and A. Badfar. 2019. Healthcare big data processing mechanisms: The role of cloud computing. International Journal of Information Management 49:271–89. doi:10.1016/j.ijinfomgt.2019.05.017.

 Web of Science ®Google Scholar

Taher, N. C., I. Mallat, N. Agoulmine, and N. El-Mawass (2019, April). An IoT-Cloud based solution for real-time and batch processing of big data: Application in healthcare. In 2019 3rd International Conference on Bio-engineering for Smart Technologies (BioSMART), Paris, France, (pp. 1–8). IEEE.

 Google Scholar

Alboaneen, D., H. Tianfield, Y. Zhang, and B. Pranggono. 2020. A metaheuristic method for joint task scheduling and virtual machine placement in cloud data centers. Future Generation Computer Systems 115:201–12. doi:10.1016/j.future.2020.08.036.

 Web of Science ®Google Scholar

Aburukba, R. O., M. AliKarrar, T. Landolsi, and K. El-Fakih. 2020. Scheduling Internet of Things requests to minimize latency in hybrid Fog–Cloud computing. Future Generation Computer Systems 111:539–2627. doi:10.1016/j.future.2019.09.039.

 Web of Science ®Google Scholar

Faraji Mehmandar, M., S. Jabbehdari, and H. Haj Seyyed Javadi. 2020. A dynamic fog service provisioning approach for IoT applications. International Journal of Communication Systems 33 (14):e4541. doi:10.1002/dac.4541.

 Web of Science ®Google Scholar

Wang, Z., D. Liu, and A. Jolfaei. 2020. Resource allocation solution for sensor networks using improved chaotic firefly algorithm in IoT environment. Computer Communications 156:91–100. doi:10.1016/j.comcom.2020.03.039.

 Web of Science ®Google Scholar

Mahini, A., R. Berangi, A. M. Rahmani, and H. Navidi. 2021. Bankruptcy approach to integrity aware resource management in a cloud federation. Cluster Computing 24 (4):3469–94. doi:10.1007/s10586-021-03336-x.

 Web of Science ®Google Scholar

Jula, A., E. A. Sundararajan, Z. Othman, and N. K. Naseri. 2021. Color revolution: a novel operator for imperialist competitive algorithm in solving cloud computing service composition problem. Symmetry 13 (2):177. doi:10.3390/sym13020177.

 Google Scholar