Multiobjective Edge Server Placement in Mobile-Edge Computing Using a Combination of Multiagent Deep <i>Q</i>-Network and Coral Reefs Optimization

Asghari, Ali; Sohrabi, Mohammad Karim

doi:10.1109/jiot.2022.3161950

Cited by 20 publications

(7 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Equation ( 4) shows the energy consumption of a processor, where 𝐸 is equal to total energy and 𝐸 𝑠 and 𝐸 𝑑 are static and dynamic energy, respectively. 𝐸 = 𝐸 𝑠 + 𝐸 𝑑 (4) The dynamic energy is shown in Equation (5). where 𝑣 ،𝑓 ،𝑃 𝑑 ،𝑡 𝑝 are the voltage, frequency, power, and time interval of processor activity, respectively.…”

Section: Energy Model and Dvfs Technicmentioning

confidence: 99%

“…Due to the feature of the PSO algorithm that is suitable for continuous problems, the authors have made changes to this algorithm to design a new model with the ability of server placement that has a discrete nature. in [5], a learning-based server placement algorithm that uses the Deep Q-Network model, and CRO algorithm (MOP-DQ) has been introduced. To reduce the time complexity of the server placement problem, the authors clustered the resource deployment area into small sub-regions.…”

Section: Compared Algorithmsmentioning

confidence: 99%

“…Proper server placement has an important role in improving service quality, and consequently, users will access the most suitable resources they need. Otherwise, server overloading and underloading may occur in different parts of the network [5]. Cloud service providers, while increasing the efficiency of their services, prefer to reduce the energy consumption of servers, which consequently reduces costs [6].…”

mentioning

confidence: 99%

See 2 more Smart Citations

Energy-aware edge server placement using the improved butterfly optimization algorithm

Asghari¹,

Sayadi²,

Azgomi

2023

J Supercomput

Self Cite

View full text Add to dashboard Cite

Cloud service providers transfer some of their resources to the proximity of their users in order to increase the quality of services provided to them. Proper placement of servers, considering the number of service demands in different parts of the network, not only plays an important role in providing better services to users but also causes more effective use of resources and reduces their energy consumption. Some related research has been done in this context. However, designing a model that can meet the needs of both the users and the service providers has received less attention. On the other hand, most researchers use discrete models to select a number of candidate locations for resource deployment, while the proposed method explores the entire search area to find optimal locations for server placement. The proposed method (ESPB) using butterfly optimization algorithm(BOA), DVFS technic, and coral reefs optimization algorithm(CRO) seeks to find the best locations for edge servers. In the first step, BOA is used to find the best locations for resource deployment. Then the CRO algorithm is used to map between the optimal locations and the servers. The experiments show that the proposed method can effectively save energy and reduces network latency.

show abstract

Section: Energy Model and Dvfs Technicmentioning

confidence: 99%

Section: Compared Algorithmsmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Energy-aware edge server placement using the improved butterfly optimization algorithm

Asghari¹,

Sayadi²,

Azgomi

2023

J Supercomput

Self Cite

View full text Add to dashboard Cite

show abstract

“…Zeng et al 26 transformed the problem of minimizing the number of edge servers while ensuring some QoS requirements into the minimum dominating set problem in graph theory, and then a greedy based algorithm is proposed to solve the problem. Recently, deep learning and reinforcement learning have also started to be applied to the edge server placement process, the deep Q‐network (DQN) and Markov game (MG) are used to optimize global resource placement to reduce global latency and to improve resource load balancing as its two objectives 27 …”

Section: Related Workmentioning

confidence: 99%

Edge server placement for offloading real‐time statistics tasks of bus passenger

Guo

Shang

et al. 2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

With the continuous development of smart bus systems, higher requirements are expected for the accuracy and timeliness of the real‐time statistics of bus passengers. Although the statistical method of image and video processing based on deep learning has higher accuracy, it has higher requirements on the computing power and hardware equipment of the computer. A traditional solution is cloud computing, but cloud computing cannot meet real‐time requirements due to long‐distance transmission. In order to meet the real‐time demand, it can be offloaded to the edge of the network and processed by edge servers. In edge computing, the location of the edge server will have a great impact on the access delay and the traffic load in the edge network. Currently, few people optimize the traffic load in the edge network during the placement process. In view of this, an edge server placement algorithm for task offloading, named ESPTO, is designed to balance the average delay and traffic load under the control of each edge server while minimizing the average delay and traffic load in the edge network. First, a decomposition‐based multi‐objective evolutionary algorithm (MOEA/D) is used to find a better set of placement strategies, and then the optimal placement strategy is obtained through TOPSIS. Experimental results based on the Hangzhou bus station dataset prove the effectiveness of ESPTO.

show abstract

“…of resources is an NP-Hard problem, some optimization methods have been used in this field. Some of these researches include PSO-based [5], GA-based [6], clustering-based [7], MIP based [8], and learning-based [9]. However, there are still important challenges.…”

mentioning

confidence: 99%

Multi-Objective Edge Server Placement Using the Whale Optimization Algorithm and Game Theory

Asghari¹,

Azgomi

Darvishmofarahi³

2022

SSRN Journal

View full text Add to dashboard Cite

Multiobjective Edge Server Placement in Mobile-Edge Computing Using a Combination of Multiagent Deep Q-Network and Coral Reefs Optimization

Cited by 20 publications

References 39 publications

Energy-aware edge server placement using the improved butterfly optimization algorithm

Energy-aware edge server placement using the improved butterfly optimization algorithm

Edge server placement for offloading real‐time statistics tasks of bus passenger

Multi-Objective Edge Server Placement Using the Whale Optimization Algorithm and Game Theory

Contact Info

Product

Resources

About