Optimized hybrid service brokering for multi-cloud architectures

Khan, Minhaj Ahmad

doi:10.1007/s11227-019-03048-5

Cited by 15 publications

(9 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, a dynamic service broker policy improved the DC selection process by using the concept of test jobs to evaluate the DCs and then used a vector space model and a multi-objective optimization technique to dynamically select the best DC (Kofahi et al, 2019). Taking into account static and dynamic parameters, a normalization-based hybrid service broker (NHSB) approach is proposed by Khan (2020). The NHSB approach considers several factors, such as the number of VMs, VM image size, VM memory, VM bandwidth, cost per VM/s, cost per VM memory, storage cost, bandwidth cost/GB, total memory, total storage, machine bandwidth and total processor speed as static parameters.…”

Section: Related Workmentioning

confidence: 99%

“…Thus, they are seeking a plan that offers running the given job at an affordable cost while maintaining an acceptable response time. Therefore, the main task of a service broker in the cloud computing paradigm is to identify and select the DC that offers the best plan in terms of improving the response time and minimizing the cost when carrying out users' jobs (Benlalia et al, 2019;Khan, 2020;Manasrah and Gupta, 2019;Youssef, 2020). Most of the service broker policies introduced in the literature focus on improving limited aspects, such as cost, response or time when running a user's job.…”

Section: Introductionmentioning

confidence: 99%

“…The second group aims to produce service broker policies to reduce the processing cost when running users' jobs (Chudasama et al, 2012;Rekha and Dakshayini, 2018;Sun et al, 2019). The third group focuses on developing and incorporating service broker policies that offer a trade-off between the response time and the processing cost when processing users' jobs (Khan, 2020;Kofahi et al, 2019;Manasrah et al, 2017;Manasrah and Gupta, 2019;Mehdi et al, 2012;Mehdi et al, 2011;Subramanian and Savarimuthu, 2016). The work presented by Arya and Dave (2017) introduces a new service broker policy for the fog computing environment that identifies the best plan for DC selection, taking into account desirable user preferences while maintaining a reasonable cost without compromising the performance of running users' jobs (Chauhan et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Efficient and Cost-effective Service Broker Policy Based on User Priority in VIKOR for Cloud Computing

Radi¹,

Alwan²,

Abualkishik³

et al. 2021

BAS-SJKFU

View full text Add to dashboard Cite

Cloud computing has become a practical solution for processing big data. Cloud service providers have heterogeneous resources and offer a wide range of services with various processing capabilities. Typically, cloud users set preferences when working on a cloud platform. Some users tend to prefer the cheapest services for the given tasks, whereas other users prefer solutions that ensure the shortest response time or seek solutions that produce services ensuring an acceptable response time at a reasonable cost. The main responsibility of the cloud service broker is identifying the best data centre to be used for processing user requests. Therefore, to maintain a high level of quality of service, it is necessity to develop a service broker policy that is capable of selecting the best data centre, taking into consideration user preferences (e.g. cost, response time). This paper proposes an efficient and cost-effective plan for a service broker policy in a cloud environment based on the concept of VIKOR. The proposed solution relies on a multi-criteria decision-making technique aimed at generating an optimized solution that incorporates user preferences. The simulation results show that the proposed policy outperforms most recent policies designed for the cloud environment in many aspects, including processing time, response time, and processing cost. KEYWORDS Cloud computing, data centre selection, service broker, VIKOR, user priorities

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Efficient and Cost-effective Service Broker Policy Based on User Priority in VIKOR for Cloud Computing

Radi¹,

Alwan²,

Abualkishik³

et al. 2021

BAS-SJKFU

View full text Add to dashboard Cite

show abstract

“…In a normal tiered fog-cloud environment, user requests are forwarded to suitable computing devices [11] based on some criteria. There are different types of user requests that are generally classified based on the type of services and task deadline.…”

Section: Introductionmentioning

confidence: 99%

“…Short deadline tasks are time-sensitive and need special resources so that they must be completed within a time while long deadline tasks can be delayed to some extent. To meet the task deadlines and with the best services, a lot of service broker techniques have been proposed to minimize the response time [11]. Most of these strategies forward the task to the nearby data centers and neglect the cost factor.…”

Section: Introductionmentioning

confidence: 99%

Cost Optimization in Cloud Environment based on Task Deadline

Ahmad

Munir

Ramzan

2022

Preprint

View full text Add to dashboard Cite

The task of a service broker in a cloud computing environment is vital and challenging. Several sophisticated techniques have been proposed to improve the overall performance and minimize the cost. Most of the approaches optimize either performance or the cost; however, the users of cloud and fog need the best compromise for both. In this paper, we propose a hybrid technique that is Cost Optimization in the cloud/fog environment based on Task Deadline (COTD) that focuses on both parameters simultaneously and maintains the Quality of Service. The proposed algorithm is based on the task deadline and selects the most appropriate data center for task processing. We have implemented and tested COTD on a well-known simulation tool Cloud Analyst. Our comprehensive testbed experiments and simulations show that COTD outperforms comparable with Service Proximity Based Routing and Performance-Optimized Routing, and successfully minimizes an average of 35% cost while maintaining the same response time.

show abstract