Mobile‐aware dynamic resource management for edge computing

Filiposka, Sonja; Mishev, Anastas; Gilly, Katja

doi:10.1002/ett.3626

Cited by 12 publications

(25 citation statements)

References 24 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They improved the previous scheme in follow-up work. [8] The authors implemented a the "Follow me" VM migration, but did not take into account the operational costs of frequent migrations. This scheme is somewhat similar to the one proposed in this paper, but we consider the problem of migration cost and reduce the invalid migration by predicting user sojourn time.…”

Section: Related Workmentioning

confidence: 99%

“…Second, we mainly dealt with the problem of migration of edge resources due to the user mobility. The first phase of initialization of VM placement problem has been studied, [8] based on which we put forward a VM migration scheme.…”

Section: The Mobility-aware Dynamic Service Placement Schemementioning

confidence: 99%

“…Service migration caused by the user mobility has been addressed in some papers. Filiposka et al [8] proposed a service migration pattern that services should be migrated among fog nodes to follow the user mobility. However, it is worth noting that frequent migration may incur additional migration costs, such as time latency and energy consumption.…”

Section: Introductionmentioning

confidence: 99%

“…Second, we construct a model of the sojourn time of the EUs at AP and the use of this model reduces invalid migration so as to improve the user's QoS. The work in this paper is based on previous studies, [8] but please note that we propose a different virtual machine (VM) migration scheme that takes into account the cost of migration. Figure 1 illustrates the scenario proposed in this paper.…”

Section: Introductionmentioning

confidence: 99%

“…Dynamic service placement in edge computing has been addressed in the previous studies. [8,9] However, we have considered some issues that were rarely noticed in previous studies. The main contributions of this paper are summarized as follows:…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Mobility-aware dynamic service placement for edge computing

Liu¹,

Wang²,

Tian³

et al. 2019

EAI Endorsed Transactions on Internet of Things

View full text Add to dashboard Cite

Edge computing provides capabilities similar to traditional cloud computing at network edge closer to users. Offloading applications from the cloud to the edge of the network can effectively diminish the latency of users waiting for applications to improve their quality of service(QoS). Due to the user mobility, however, the distance between the user and the edge device is constantly increasing, resulting in a degradation of the user's QoS. To satisfy the QoS of mobile users, applications should be dynamically migrated to follow the user mobility.In this paper, we propose a mobility-aware dynamic service placement policy that reduces the number of virtual machine migrations and diminishes user-perceived latency simultaneously. The proposed scheme filters out the invalid migration that the user handoffs to another access point before the migration is completed. The simulation trials demonstrate that our solution improves the QoS of users and reduces the number of application migrations.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: The Mobility-aware Dynamic Service Placement Schemementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Mobility-aware dynamic service placement for edge computing

Liu¹,

Wang²,

Tian³

et al. 2019

EAI Endorsed Transactions on Internet of Things

View full text Add to dashboard Cite

show abstract

Efficient data transfer in edge envisioned environment using artificial intelligence based edge node algorithm

Kumar

Batth

et al. 2020

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Edge computing technology has drawn the keystone of future intelligent transportation systems, especially in smart cities, because of processing data that are near to the user location present at the edge of the cloud server. Generally, in smart cities where distributed things have access to computational resources, data transfer becomes inevitable because of high latency, thereby resulting in crucial situations. Even though numerous technologies have emerged for improving the data communication among geo-distributed devices received from the cloud server but still, it lags in low learning performance. To address these challenges, an innovative artificial intelligence (AI) based edge node (E-Node) algorithm is implemented to optimize the edge to edge learning for well-organized data migration. To attain high reliability, AI-K-means neural network (KNN) and convolution neural network (CNN) is used initially for pre-processing and filtering the edge node. Further, the proposed E-Node algorithm outperforms the optimization technique effectively through the edge to edge computing The reliability performance is increased by reducing the node optimization time from 132 to 98 ms for 25 kb data and data transmission time is reduced from 93 to 45 ms for 80 kb data, thus reducing the latency in an edge envisioned environment. INTRODUCTIONArtificial intelligent (AI) frameworks help to solve problems involving the learning abilities of computers, language processing, and identifying various speech in the environment in a highly reliable and accurate manner. 1 Smart cities concept endures the cities with intelligence and other concepts involving smart vehicles, smart grid, and smart healthcare using the concept of deep learning. Internet of Vehicles (IoV) 2,3 lays the foundation for several new transportation systems . 4 IoV helps to improve the efficiency of transportation, reducing the accident rate and reducing the energy consumption in vehicles being reduced. 5,6 The data generated from these smart vehicles are stored in the cloud. However, it is difficult to handle several request messages from the client and process it, during a particular grave learning task. The proposed system uses a new model to foil centralized cloud computing, which reduces the latency during communication and improves the quality through the provision of various resources 7 that are near to terminal devices for latency-sensitive tasks. The use of edge to edge cooperative AI increases the learning efficiency, quality of service. It also improves variousAbbreviations: CNN, convolution neural network; E-Node, edge node; KNN, K-means neural network.

show abstract