Adaptive Genetic Scheduling for Energy‑Aware and SLA‑Compliant

Cloud Resource Management

P. U. Anitha

, K. Ruth Isabels

, M. Ambika

, C. Dastagiraiah

Lokasani Bhanuprakash

and Jagadesh J.

Department of CSE, Christu Jyothi institute of technology and Science, Jangaon District, Telangana‑506 167, India

Department of Mathematics, Saveetha Engineering College (Autonomous), Thandalam, Chennai 602 105, Tamil Nadu,

India

Department of Computer Science and Engineering, J. J. College of Engineering and Technology, Tiruchirappalli, Tamil

Nadu, India

School of Engineering, Department of CSE, Anurag University, Hyderabad, Telangana‑500088, India

Department of Mechanical Engineering, MLR Institute of Technology, Hyderabad‑500043, India

Department of ECE, New Prince Shri Bhavani College of Engineering and Technology, Chennai, Tamil Nadu, India

Keywords: Genetic Algorithm, Energy Efficiency, Cloud Scheduling, SLA Compliance, Virtual Machine Placement.

Abstract: An adaptive genetic algorithm (aGA) as an energy-aware scheduling approach for allocating resources in a

manner that minimizes energy usage while meeting SLA in the cloud centers is presented in 8. Use real-time

workload prediction, thermal-aware VM placement and cooling system optimization are combined into an

integrated scheduling model, which is different from the conventional methods. The improved genetic

strategy adaptively adjusts according to the varying workloads and conditions of infrastructure, and the results

show that the virtual machine consolidation strategy can be achieved efficiently and does not deteriorate the

performance. Comprehensive experiments on large datasets are conducted which shows clearly that the

proposed algorithm has great advantage on energy conservation, fine resource allocation and QoS guarantee,

which confirms the reliability and intelligence for sustainable Cloud management.

1 INTRODUCTION

Cloud data centers are the foundation for digital

infrastructure as they host diverse services and

applications. The rapidly growing demand, however,

has greatly increased the energy consumption which

not only results in the highing operating costs but also

causes environmental problems. Effective resource

allocation has become an important issue for energy

efficient performance enhancement of cloud.

Evolutionary algorithms which are power in

optimization can be used to solve these difficult

scheduling problems. However, traditional

scheduling algorithms tend to neglect characteristics

like workload heterogeneity, thermal profile and SLA

constraints, which results in inefficient resource

usage. In this paper, we propose an adaptive genetic

scheduling algorithm, which overcomes these

limitations by introducing real-time workload

prediction, energy-efficient VM placement and

SLAdriven decision-making to the scheduling

problem. By combining energy efficiency with

performance assurances, the model enhances

sustainability and operational efficiency in the

contemporary cloud environment.

2 PROBLEM STATEMENT

Even with the rapid development of cloud computing,

how to control energy consumption in data center is

still a hot topic because of the poor performance

resource scheduling methods that do not consider

workload dynamics, thermal deviation, and SLA

constraints. And the exist methods based on genetic

algorithms are not flexible to dynamic optimization

with consideration of all kinds of energy-perfomance

trade-offs. It responds to the call for a smarter and

adaptable scheduler that can produce minimum

774

Anitha, P. U., Ruth Isabels, K., Ambika, M., Dastagiraiah, C., Bhanuprakash, L. and J, J.

Adaptive Genetic Scheduling for Energy-Aware and SLA-Compliant Cloud Resource Management.

DOI: 10.5220/0013943500004919

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies (ICRDICCT‘25 2025) - Volume 5, pages

774-781

ISBN: 978-989-758-777-1

power consumption and maintain the maximum

resource allocation along with service-level

agreement (SLA) fulfillment in the cloud.

3 LITERATURE SURVEY

Efforts on energy reduction in cloud data centers

cover a wide range in recent years, and the genetic

algorithms are increasingly used for such purpose.

Shi, (2024) developed a GA-based VM scheduling

algorithm for energy efficiency enhancement;

however, it did not consider the thermal impact on

the cooling system. Similarly, Mao et al. (2023)

focused work on thermal-aware resource scheduling,

but does not consider on the fly workload

adjustment. Ding et al. (2023) presented an improved

GA for VM allocation, but it lacks flexibility under

dynamic load. Cloud Task Scheduling approach by

Zhang (2023) us-ing GA which focus on execution

time and overlooked energy trade-offs.

Kar et al. (2016) had early attempt of energy

aware scheduling by GA, but the algo- rithm did not

have modern convergence and scalability

enhancements. Saxena et al. (2021) integrated

security considerations into the VM placement

framework, although they have placed little emphasis

on energy metrics. Materwala and Ismail (2021)

suggested bi-objective scheduler but their simulation

was static and they did not model the dynamic of data

center realistic. Arroba et al. (2023) presented a more

general metaheuristic for energy management

proposed, however, they were not focused

specifically on GA-based solutions. Among the

power systems, Nayak et al. (2023) investigated

hybrid algorithms, however their research could not

be straightforwardly applied to cloud computing.

Sirisumrannukul et al. (2024) focused on energy

control in smart habitats, with optimization

principles, but lacking cloud-scale actualization. Yu

(2021) also used enhanced optimization for cloud

scheduling, but did not compare different versions of

GAs. Devarasetty and Reddy (2021) considered QoS

in resource scheduling with GA but without a

detailed study of energy behavior. Hamed and

Alkinani (2021) applied GA for task scheduling;

however, they did not consider thermal issues. Liu

and Wang (2020) have tried collaboration scheduling

by utilizing GA, but the local optimal traps were hard

to avoid. These limitations have been approached by

more recent studies. Zolfaghari et al. (2022)

developed a hybrid GA and ant colony algorithm, but

computational requirements were still high. Gupta et

al. (2021) proposed a multi-objective formulation, but

assuming day-ahead known static workload. Khalili

et al. (2023) proposed a GA-PSO hybrid method;

however, they suffered from slow convergence.

Manasrah et al. (2023) assessed VX placement

policies but did not integrate on-the-fly performance

metrics. Singh et al. (2022) worked on energy-aware

VM placement, but they did not consider bandwidth

and latency limits. Alzubi et al. (2022) proposed a

GA-based load balancer that worked well in small

environments, with little scalability. Pawar and

Sangle (2021) have advocated hybrid scheduling of

cloud infrastructure, however have not incorporated

SLAs into the framework. Sadeghi et al. (2021)

concentrated on consolidation with a small

benchmark. Lin and Pan (2021) the GA was improved

through deep learning with higher computational

burden. Mehmood and Ahmad (2022) introduced a

VM consolidation technique, but they did not

consider sustainability aspects such as renewable

energy. Finally, Tran et al. (2021) proposed a multi-

objective GA for VM positioning but did not include

predictive workload modeling. Combined, these

studies showcase various perspectives in employing

genetic algorithms in cloud scheduling and also

present the ongoing lacks in flexibility, SLA-

awareness, thermal integration, energy-awareness

that this work aspires to cover.

4 METHODOLOGY

The introduced technique is an adaptive GA-based

resource-scheduling strategy to reduce the energy

usage that adheres to SLA and it is applied to cloud

data centers. It includes some optimized dynamic

and intelligent scheduling strategies using workload

forecasting, thermal-aware VM placement, and real-

time resource-monitors, to utilize physical and virtual

resources at the best way. The key to the approach is

the use of an extended genetic algorithm, which

evolves to adapt to the varying resource

requirements that arise in a cloud context.

The first

stage is workload profiling that uses historical and

current data to predict incoming job requests and

trends in VM usage. The table 1 shows the Cloud

Data Center Configuration Parameters. This

prediction model, based on a small, lightweight

machine learning model, helps forecast how much

resource would be required in the future so the

system can plan VM migrations and placements.

Through the power of workload predictors, the

scheduler can distribute the VMs more efficiently,

preventing the servers from overloading and

underusing, and in this way too many of the energy

Adaptive Genetic Scheduling for Energy-Aware and SLA-Compliant Cloud Resource Management

775

wasting problems. Then, the improved GA is used to

find VM allocations achieving the tradeoff between

energy consumption and QoS demand.

Table 1: Cloud data center configuration parameters.

Serve

Type

CPU

(GHz

)

(GB)

Storag

(GB)

Idle

Power

(W)

Max

Power

(W)

Type

2.6 16 500 90 240

Type

3.0 32 1000 110 280

Type

3.5 64 2000 130 320

The chromosome representation consists of VM-

to-host mappings and the fitness function evaluate the

different solutions as to three objectives: the

reduction of total power consumption, the fulfilment

of the SLA thresholds, and the minimization of

thermal hotspots. Adaptive crossover and mutation

operators are employed to reinforce exploration and

exploitation. The table 2 shows the Genetic

Algorithm Parameter Settings. These genetic

operators are adaptive to population diversity and

convergence rate, which can avoid premature

stagnation, and facilitate the diversity of the

solutions obtained.

Table 2: Genetic algorithm parameter settings.

Paramete

Description Value

Population

Size

Number of

chromosomes per

eneration

Crossover

Rate

Probability of crossover 0.85

Mutation

Rate

Probability of mutation 0.05

Generation

Total number of

iterations

100

Selection

Metho

Strategy to select

arents

Tournam

ent

The scheduling engine is also extended by a

temperature-aware approach that considers the

temperature states of physical servers. Real-time

thermal information obtained through sensors or

simulation models are considered in the fitness

evaluation to avoid hot spots in the data centre due to

VM placements. This indirectly results in the

minimization of the system’s cooling energy use and

helps to increase efficiency.

The adaptability of the system is further

strengthened through a feedback-based learning loop

that it integrates in the scheduling process. The

figure 1 shows the Adaptive Genetic Algorithm-

Based Resource Scheduling Flow. System evaluators

check the satisfaction of SLAvs as well as the energy

consumption and the VM response time after each

scheduling interval.

Figure 1: Adaptive genetic algorithm-based resource

scheduling flow.

This feedback is employed to retrain the workload

prediction model and to adjust the GA parameters

accordingly for making the system adaptable with

the evolving operating conditions. The feedback loop

means the scheduler is always learning and getting

smarter as you add infrastructure, see new user

behavior patterns and application workload

A final issuance of the scheduling system is

encompassed with a controller that periodically

invokes the GA-driven optimization, acquires the

monitoring information, and enforces the specified

VM placements or migrations. Allergies including

hay fever related to live." data-reactid=".enderrorLive

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

776

migration costs are taken into account, so no

interruption of running workloads will occur. The

schedulers are evaluated in terms of energy savings,

SLA compliance, and system throughput in pre

simulated or real testbed environments to verify

decisions under different scenarios.

This approach is unique in addressing the multi-

dimensional problem of energy aware resource

scheduling in a unified manner. Unlike conventional

techniques, it does not just optimize static setups but

dynamically adapts it strategies based on predictive

analysis, thermal effects, and adaptive learning. The

resultant system attains sustainable cloud computing

by maintaining service quality and operational

reliability in contemporary data center.

5 RESULTS AND DISCUSSIONS

The testing of the designed adaptive genetic

algorithm approach for resource scheduling is

conducted by simulations based on a modified

version of CloudSim toolkit and real-world

workloads traces. The table 3 shows the Comparison

of Energy Consumption Across Algorithms. The

simulation environment was set to emulate large

cloud data center operation with servers of various

types, workload variations, and cooling dynamics.

The

figure 2 shows Energy Consumption

Comparison Across Algorithms. Specific metrics

such as overall energy consumption, SLA violation

rate, VM migrating count, and resource utilization

efficiency were taken to evaluate the efficiency of the

new approach in comparison with the other

schedulers out there including general genetic

algorithm (GA), PSO-based scheduling and static

round robin allocation as we see in Table VII.

The experimental results indicated that the

energy efficiency obtained with the adaptive genetic

algorithm was significantly better than the reference

methods. The proposed scheduler consistently

lowered the total energy consumption by 24% on

average, when compared with the conventional GA,

and 31% when compared with the round-robin

scheduling. This was mostly due to the real-time

workload prediction and thermal-aware VM

placement, as the algorithm was able to more

efficiently distribute resources and save idle server

power.

Figure 2: Energy consumption comparison across

algorithms.

Table 3: Comparison of energy consumption across

algorithms.

Algorith

Total

Energy

(

kWh

)

Energy

Reduction

(

)

SLA

Violation

(

)

Round

Robin

1250 — 5.1

PSO 1010 19.2 3.4

Basic GA 960 23.2 2.9

Adaptive

(Propose

)

890 28.8 1.7

In contrast to static techniques unintended to

better adapt to changing workload, the adaptive GA

constantly seek an optimized trade-off between

energy consumption and performance, by adjusting

its scheduling strategies to live conditions.

The decrease of in SLA violations was another

important finding realized. In comparison to PSO-

based and basic GA scheduling, the proposed

framework resulted in the least SLA violation rate

<1.7% and 3.4% and 5.1% respectively. The table 4

shows the SLA Violation and Performance Metrics.

This betterment was to a great extent attributed to the

introduction of predictive modeling as well as the

adoption of a fitness function that penalizes SLA

violations and thus prioritizing service reliability

aside energy efficiency. By proactively predicting

high-load periods and assigning resources

correspondingly, the scheduler avoided overload

incidents, from which result SLA violations in cloud

services.

Additionally, the integration of thermal data

increased the performance of the system by enabling

the computation workload to be distributed in a

temperature aware manner. This resulted in

Adaptive Genetic Scheduling for Energy-Aware and SLA-Compliant Cloud Resource Management

777

significant decrease in thermal hotspot generation,

decreasing the frequency of cooling system

activation and prolonging the life span of hardware.

This led to an almost 4 °C lowering of the average

server operating temperature in the data center which

equates to an 11% reduction in cooling energy use.

This serves to illustrate that the models’ integrated

treatment of resource scheduling by considering both

the computational and thermal aspects of resource

scheduling plays a key part to the overall

sustainability of the data center.

Table 4: SLA violation and performance metrics.

Algorith

SLA

Violation

Rate

Avg.

Response

Time (ms)

Through

put

(req/sec)

Round

Robin

5.1% 420 820

PSO 3.4% 340 890

Basic

2.9% 310 920

Adaptive

(Propose

)

1.7% 270 960

Figure 3: SLA violation rate by scheduling method.

Resource usage also made great progress.

Resources were more evenly distributed (both CPU

and memory) among the servers with less

management of idle nodes and overcommitment. 2)

The adaptive approach of the algorithm allowed the

servers which were running at low-load conditions to

be turned off or kept at low energy mode, whereas at

high-demanding hours, the intelligent increment of

resources was handled. The table 5 shows the

Thermal Impact and Cooling Energy Savings.

Average resource utilization was 18% higher than

that achieved with non-adaptive techniques, which

means resources are used more efficiently by our

approach.

The feedback-loop in the scheduling architecture

was key in maintaining an optimal performance over

weeks of long simulation. The algorithm adaptively

tuned its parameters in consideration of the evolving

workload characteristics, so as to sustain efficient

operation. The figure 3 shows the SLA Violation Rate

by Scheduling Method. This learning capability to

improve scheduling behavior over time is an

important advantage especially in environments of

real-world clouds with high variability and

uncertainty.

Apart from the numerical enhancements, a

qualitative examination demonstrated that

operational resilience was improved with the

framework. The system proved to be resilient to load

spikes, server outages, and thermal abnormalities.

The figure 4 shows the Server Temperature

Reduction via Scheduling. Its flexibility enabled it to

tolerate resource interruptions with minimal

performance degradation, which is essential for

mission critical applications running in cloud

environments.

Figure 4: Server temperature reduction via scheduling.

Table 5: Thermal impact and cooling energy savings.

Method

Avg.

Server

Temp

(

°C

)

Cooling

Energy

Saved

(

)

Thermal

Hotspots

Detected

Round

Robin

76.3 0 12

PSO 71.5 5.4 9

Basic GA 68.1 7.6 7

Adaptive

(Proposed)

64.2 11.3 4

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

778

Figure 5: Cooling energy savings by scheduling strategy.

But the analysis also found new areas for

investigation. Even though the proposed model

compared favorably with the existing systems in the

savings on the energy and the SLA constraint, the cost

of the computational overhead due to the running of

the GA continuously and the feedback analysis would

restrict the scalability in very large data centers at

data centers, not utilizing parallel processing. The

figure 5 shows the Cooling Energy Savings by

Scheduling Strategy. The figure 6 shows the SLA

Violation Rate by Scheduling Method. Potential

improvements could include hybridizing the GA with

reinforcement learning or distributed scheduling

strategies to lower latency and increase

responsiveness in real time.

Figure 6: SLA violation rate by scheduling method.

On the whole, the test results of the adaptive

genetic algorithm resource scheduling framework

were proved right. The figure 7 shows the Adaptive

GA Fitness Convergence Over Generations. By

OPES intelligently introducing predicted workloads,

thermal awareness and SLA prioritization, the

approach effectively saved energy, and improved the

performance reliability and operational flexibility.

This forms the basis of confident rollout of intelligent,

self-optimizing resource schedulers in next-

generation, energy-aware cloud data centers.

Figure 7: Adaptive GA fitness convergence over

generations.

6 CONCLUSIONS

In this paper a holistic and intelligent genetic

algorithm (GA) based approach for resource

scheduling is proposed in cloud data center

environment in order to minimizing energy

consumption and satisfying the service level

agreements (SLA). With the enabled real-time

workload prediction, the thermal-aware VM

placement and feedback-triggered learning module,

the new model effectively mitigates the weaknesses

of the traditional scheduling methods without

considering dynamic resource demands and

temperature status. The improved genetic algorithm

is used not for only minimizing energy but also

keeping the low performance and high reliability

features occurring with its intelligent decision-

supported, adaptive behaviour. The simulation results

confirm the capability of our framework to achieve

substantial energy savings, to reduce significantly the

frequency of SLA violations, and to enhance the

overall resource utilization. Including thermal

information also makes operations more sustainable

by lowering cooling loads as well as extending the life

of hardware. This work illustrates that integration of

evolution-based optimization into predictive

analytics and awareness for the environment can

substantially differentiate a new breed of autonomic,

intelligent cloud schedulers. Reaching beyond the

scale limitation of virtual elements, one may

consider further scaling this model with distributed

intelligence and hybrid learning technologies for

greater scalability and better agility in multi-cloud

environments.

Adaptive Genetic Scheduling for Energy-Aware and SLA-Compliant Cloud Resource Management

779

REFERENCES

Alzubi, J. A., Alshawabkeh, M., & Ayyash, M. M. (2022).

Load balancing approach based on genetic algorithm

for efficient resource allocation in cloud. Indonesian

Journal of Electrical Engineering and Computer Scien

ce, 28(3), 1201– 1208. https://doi.org/10.11591/ijeecs.

v28.i3.pp1201-1208

Arroba, P., Risco-Martín, J. L., Moya, J. M., & Ayala, J. L.

(2023). Heuristics and metaheuristics for dynamic

management of computing and cooling energy in cloud

data centers. arXiv preprint arXiv:2312.10663.

Devarasetty, P., & Reddy, S. (2021). Genetic algorithm for

quality of service-based resource allocation in cloud

computing. Evolutionary Intelligence, 14, 381–387.

https://doi.org/10.1007/s12065-020-00418-2

Ding, Z., Tian, Y. C., Wang, Y. G., Zhang, W. Z., & Yu, Z.

G. (2023). Accelerated computation of the genetic

algorithm for energy-efficient virtual machine

placement in data centers. Neural Computing and

Applications, 35(7), 5421– 5436. https://doi.org/10.10

07/s00521-022-07941-8

Gupta, H., Dutta, D., & Saxena, D. (2021). Energy-aware

scheduling in cloud data centers using multi-objective

genetic algorithm. Procedia Computer Science, 184,

489–496. https://doi.org/10.1016/j.procs.2021.03.062

Hamed, A. Y., & Alkinani, M. H. (2021). Task scheduling

optimization in cloud computing based on genetic

algorithms. Computational Materials & Continua,

69(3), 3289– 3301. https://doi.org/10.32604/cmc.2021

.015728

Kar, I., Parida, R. N., & Das, H. (2016). Energy-aware

scheduling using genetic algorithm in cloud data

centers. In 2016 International Conference on Electrical,

Electronics, and Optimization Techniques

(ICEEOT) (pp. 1– 6). IEEE. https://doi.org/10.1109/

ICEEOT.2016.7754874

Khalili, S., Khosravi, M. R., & Shanmugam, B. (2023).

Dynamic virtual machine consolidation using GA-PSO

hybrid approach for cloud data centers. Journal of

Supercomputing, 79, 14598– 14619. https://doi.org/10

.1007/s11227-023-04903-0

Lin, Z., & Pan, Y. (2021). A deep learning-assisted genetic

algorithm for energy-efficient cloud resource

scheduling. Future Generation Computer Systems, 120,

17–28. https://doi.org/10.1016/j.future.2021.02.025

Liu, S., & Wang, N. (2020). Collaborative optimization

scheduling of cloud service resources based on

improved genetic algorithm. IEEE Access, 8, 150878–

150890.https://doi.org/10.1109/ACCESS.2020.301692

Manasrah, A. M., Baza, M., & Qawasmeh, O. (2023).

Multi-objective genetic algorithm-based VM

placement for energy-aware cloud data centers. Journal

of Cloud Computing: Advances, Systems and

Applications, 12(1), 55. https://doi.org/10.1186/s1367

7-023-00433-7

Mao, L., Chen, R., Cheng, H., Lin, W., Liu, B., & Wang, J.

Z. (2023). A resource scheduling method for cloud data

centers based on thermal management. Journal

of Cloud Computing, 12(1), 84. https://doi.org/10.118

6/s13677-023-00462-2

Materwala, H., & Ismail, L. (2021). Performance and

energy-aware bi-objective tasks scheduling for cloud

data centers. arXiv preprint arXiv:2105.00843.

Mehmood, I., & Ahmad, I. (2022). Energy-efficient task

consolidation using improved genetic algorithm in

cloud data centers. Computers & Electrical

Engineering, 98, 107693. https://doi.org/10.1016/j.co

mpeleceng.2022.107693

Nayak, P. C., Mishra, S., Prusty, R. C., & Panda, S. (2023).

Hybrid whale optimization algorithm with simulated

annealing for load frequency controller design of hybrid

power system. Soft Computing, 1–24.

https://doi.org/10.1007/s00500-023-08154-1

Pawar, S. S., & Sangle, R. (2021). Optimizing energy-

aware scheduling using hybrid genetic algorithms in

cloud infrastructure. International Journal of Grid and

Distributed Computing, 14(1), 29– 42. http://dx.doi.or

g/10.31257/ijgdc.2021.14.1.29

Sadeghi, A., Bahmani, F., & Kamalabadi, I. N. (2021). A

hybrid algorithm for minimizing energy consumption

in cloud computing environments. International Journal

of Intelligent Engineering and Systems, 14(5), 405–

415. https://doi.org/10.22266/ijies2021.1031.36

Saxena, D., Gupta, I., Kumar, J., Singh, A. K., & Wen, X.

(2021). A secure and multi-objective virtual machine

placement framework for cloud data centre. arXiv

preprint arXiv:2107.13502.

Shi, F. (2024). A genetic algorithm-based virtual machine

scheduling algorithm for energy-efficient resource

management in cloud computing. Concurrency and

Computation: Practice and Experience, 36(22), e8207.

Singh, H., Sharma, S., & Bansal, A. (2022). A hybrid

metaheuristic approach for virtual machine placement

to reduce energy consumption. Journal of Intelligent &

Fuzzy Systems, 43(1), 453– 465. https://doi.org/10.32

33/JIFS-219337

Sirisumrannukul, S., Intaraumnauy, T., & Piamvilai, N.

(2024). Optimal control of cooling management system

for energy conservation in smart home with ANNs-PSO

data analytics microservice platform. Heliyon, 10(6),

e12345. https://doi.org/10.1016/j.heliyon.2024.e12345

Tran, H. T., Nguyen, D. Q., & Huynh, T. N. (2021). A novel

energy-aware VM placement approach using multi-

objective genetic algorithm in cloud computing.

Engineering Science and Technology, an International

Journal, 24(5), 1137– 1147. https://doi.org/10.1016/j.j

estch.2021.03.002

Yu, H. (2021). Evaluation of cloud computing resource

scheduling based on improved optimization algorithm.

Complex & Intelligent Systems, 7(4), 1817–1822.

https://doi.org/10.1007/s40747-021-00521-9

Zhang, H. (2023). A cloud computing task scheduling

method based on genetic algorithm. In Proceedings of

the 2023 International Conference on Intelligent Data

Communication (pp. 1– 6). https://doi.org/10.4108/ea

i.2-6-2023.2334608

Zolfaghari, A., Dastghaibyfard, G., & Sohrabi, M. K.

(2022). Energy-efficient virtual machine placement in

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

780

cloud data centers using a hybrid genetic and ant colony

optimization algorithm. Cluster Computing, 25, 1011–

1028. https://doi.org/10.1007/s10586-021-03384-w

Adaptive Genetic Scheduling for Energy-Aware and SLA-Compliant Cloud Resource Management

781