Optimal container resource allocation in cloud architecture: A new hybrid model

Vhatkar, Kapil N.; Bhole, Girish P.

doi:10.1016/j.jksuci.2019.10.009

Cited by 19 publications

(13 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different experiments have been carried out, particularly, experiment 1 was done using 250 machines with a capacity of 100, experiment 2 was done using 300 machines with a capacity of 200, experiment 3 was done using 350 machines with a capacity of 400 and experiment 4 was done using 400 machines with capacity 800”. Furthermore, the adopted approach was calculated to the traditional schemes like velocity updated gray wolf optimization (VU‐GWO), 52 whale random update assisted lion algorithm (WR‐LA), 53 customized rider optimization algorithm (C‐ROA) 54 and MFO 45 and the outcomes were examined in terms of statistical analysis.…”

Section: Resultsmentioning

confidence: 99%

Self‐improved moth flame for optimal container resource allocation in cloud

Vhatkar¹,

Bhole²

2022

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

SUMMARY Resource allocation in the cloud is becoming more complicated and challenging due to the rising necessities of cloud services. Effective management of virtual resources in the cloud is of large significance since it has a great impact on both the operational cost and scalability of the cloud environment. Nowadays, containers are becoming more popular in this regard due to their characteristics like reduced overhead and portability. Conventional resource allocation schemes are usually modeled for the migration and allocation of virtual machines (VM), as a result; the question may arise on, “how these strategies can be adapted for the management of a containerized cloud”. This work evolves the solution to this issue by introducing a new fitness oriented moth flame algorithm (F‐MFA) for optimizing the allocation of containers. Further in this work, the optimal allocation relies on certain constraints like balanced cluster use, system failure, total network distance (TND), security and threshold distance, and credibility factor as well. In the end, the supremacy of the presented model is computed to the conventional models in terms of cost and convergence analysis.

show abstract

Section: Resultsmentioning

confidence: 99%

Self‐improved moth flame for optimal container resource allocation in cloud

Vhatkar¹,

Bhole²

2022

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Further, the analysis was done for four experiments in terms of statistical analysis and cost function and the outcomes were obtained. At 80th iteration, the performance of the adopted model for experiment 1 has attained an improved solution with reduced cost over GA, SW-GA, SH-GA, GM-GA, LA, WOA, WR-LA, PSO and GWO schemes by 26 GWO models respectively. Thus the betterment of the adopted VU-GWO scheme has been confirmed from the simulation outcomes.…”

Section: Resultsmentioning

confidence: 99%

“…Here, experiment 1 was carried out by deploying 250 machines with capacity 100, experiment 2 was carried out with 300 machines and capacity 200, experiment 3 was carried out by deploying 350 machines with capacity 400 and experiment 4 was carried out by deploying 400 machines with capacity 800. Also, the performance of the implemented VU‐GWO technique was examined and compared over other existing schemes namely; GA [32], swap mutation‐based GA (SW‐GA) [6], shrink mutation‐based GA (SH‐GA) [6], growth mutation‐based GA (GM‐GA) [6], lion algorithm (LA) [33] and whale optimisation algorithm (WOA) [34] whale random update assisted LA (WR‐LA) [35], PSO [30] and GWO [31] and the results were achieved in terms of cost function and statistical analysis.…”

Section: Resultsmentioning

confidence: 99%

“…The importunate issue in data centers for cloud suppliers is unmistakably presented on over‐provisioning, in higher operating cost and incompetent resource utilisation may happen in the resource wastage. An archetypal data center operates with lower effectiveness, while the owners of application are yet alleged for unexploited deployed resources [26]. Under such usage is made conceivable by another explanation that has the attribute to resource acquisition's coarse granularity utilising virtual machines.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Particle swarm optimisation with grey wolf optimisation for optimal container resource allocation in cloud

Vhatkar¹,

Bhole²

2020

IET Networks

Self Cite

View full text Add to dashboard Cite

In the cloud sector, as the applications used by users are exploited via micro-service pattern, the container allocation seems to be the most vital process. This has further been concentrated with more care for its beneficiary acts like easier employment, limited overheads and higher portability. For the past few decades, various contributions have been made under the container management and allocation as well. Under these circumstances, this study intends to design an optimal resource allocation and management model by incorporating the concept of optimisation, which guarantees optimal container allocation. To make this possible, this study establishes a novel hybrid algorithm, namely velocity-updated grey wolf optimisation (VU-GWO), which is the hybridisation of two renowned algorithms particle swarm optimisation and grey wolf optimization, respectively. More importantly, the solution of optimised resource allocation is influenced by the designing of a novel objective function, which concerns the constraints like balanced cluster use, threshold distance, system failure, and total network distance as well. At last, the performance of the presented scheme is evaluated over other traditional schemes, and the betterment of the proposed model is validated.

show abstract

“…In this, Experiment 1 was exploited by using 250 machines with capacity 100, Experiment 2 was exploited with 300 machines and capacity 200, Experiment 3 was exploited by deploying 350 machines with capacity 400 and Experiment 4 was exploited by deploying 400 machines with capacity 800. Additionally, the implemented approach in terms of performance was differentiated over other traditional structures such as ROA (Binu and Kariyappa, 2018), whale optimisation algorithm (WOA) (Mirjalili and Lewis, 2016), whale random update assisted LA (WR-LA) (Vhatkar and Bhole, 2019), particle swarm optimisation (PSO) (Zhang and Xia, 2017), grey wolf optimisation (GWO) (Mirjalili et al , 2014) and velocity updated grey wolf optimisation (VU-GWO) (Netaji and Bhole, 2020) and the outcomes were analyzed. Table 2 depicts the diverse parameters of algorithms.…”

Section: Resultsmentioning

confidence: 99%

Improved rider optimization for optimal container resource allocation in cloud with security assurance

Vhatkar¹,

Bhole²

2020

IJPCC

Self Cite

View full text Add to dashboard Cite

Purpose The containerization application is one among the technologies that enable microservices architectures, which is observed to be the model for operating system (OS) virtualization. Containers are the virtual instances of the OS that are structured as the isolation for the OS atmosphere and its file system, which are executed on the single kernel and a single host. Hence, every microservice application is evolved in a container without launching the total virtual machine. The system overhead is minimized in this way as the environment is maintained in a secured manner. The exploitation of a microservice is as easy to start the execution of a new container. As a result, microservices could scale up by simply generating new containers until the required scalability level is attained. This paper aims to optimize the container allocation. Design/methodology/approach This paper introduces a new customized rider optimization algorithm (C-ROA) for optimizing the container allocation. The proposed model also considers the impact of system performance along with its security. Moreover, a new rescaled objective function is defined in this work that considers threshold distance, balanced cluster use, system failure, total network distance and security as well. At last, the performance of proposed work is compared over other state-of-the-art models with respect to convergence and cost analysis. Findings For experiment 1, the implemented model at 50th iteration has achieved minimal value, which is 29.24%, 24.48% and 21.11% better from velocity updated grey wolf optimisation (VU-GWO), whale random update assisted LA (WR-LA) and rider optimization algorithm (ROA), respectively. Similarly, on considering Experiment 2, the proposed model at 100th iteration attained superior performance than conventional models such as VU-GWO, WR-LA and ROA by 3.21%, 7.18% and 10.19%, respectively. The developed model for Experiment 3 at 100th iteration is 2.23%, 5.76% and 6.56% superior to VU-GWO, WR-LA and ROA. Originality/value This paper presents the latest fictional optimization algorithm named ROA for optimizing the container allocation. To the best of the authors’ knowledge, this is the first study that uses the C-ROA for optimization.

show abstract

Optimal container resource allocation in cloud architecture: A new hybrid model

Cited by 19 publications

References 27 publications

Self‐improved moth flame for optimal container resource allocation in cloud

Self‐improved moth flame for optimal container resource allocation in cloud

Particle swarm optimisation with grey wolf optimisation for optimal container resource allocation in cloud

Improved rider optimization for optimal container resource allocation in cloud with security assurance

Contact Info

Product

Resources

About