


Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
IT POINTS THE ADVANTAGES OF FOG OVER CLOUD
Typology: Thesis
1 / 4
This page cannot be seen from the preview
Don't miss anything!
IEEE-ICASI 2017 - Meen, Prior & Lam (Eds)
12a*
3b*
4c
5d
2e
2f
(^1) Computer and Information Networking Center, National Taiwan University, Taipei, Taiwan (^2) Department of Electrical Engineering, National Taiwan University of Science & Technology, Taipei, Taiwan (^3) Department of Computer Science and Information Engineering, National Taitung University, Taitung, Taiwan (^4) Department of Sport and Health Promotion, TransWorld University, Yunlin, Taiwan (^5) Department of Information Science and Management Systems, National Taitung University, Taitung, Taiwan achenyc@ntu.edu.tw, bycc@nttu.edu.tw, c (^) chris@twu.edu.tw, d (^) ysl@nttu.edu.tw, e (^) lchen1215@gmail.com, f (^) changbilly246@gmail.com *Corresponding author: chenyc@ntu.edu.tw, ycc@nttu.edu.tw
Abstract
This work develops a cloud-fog computing architecture for information-centric Internet-of-Things (IoT) applications with job classification and resource scheduling functions. The two designed functions support quality of service (QoS) by classifying IoT applications and scheduling computing resources. An innovative scheduling mechanism is developed to optimize the dispatch of cloud and fog resources at minimum cost in a cloud-fog computing environment. One cloud, three fogs, four types of IoT application and 2500 IoT tasks are designed in the cloud-fog simulation environment. The simulation results reveal that the proposed computing architecture outperforms computing without scheduling by reducing the cloud computing load by approximately 20.47%. The QoS of real-time IoT applications is guaranteed in the proposed mechanism
Key words: Internet-of-Things (IoT), Cloud-Fog Computing, Quality-of-Service (QoS), Information-Centric Networking
Introduction
With the rapid development of the Internet, cloud computing has emerged as an important issue. Cloud computing allows developers to deploy resources. Fog computing is referred to as โedge computingโ, which essentially means that rather than being hosted from a centralized cloud [1]. That concentration enables data to be processed locally in smart devices rather than being sent to the cloud for processing. In information-centric networking (ICN) architecture, who is communicating is less significant than that data that are required. The shift toward information-centric fog computing has occurred as a result of end-usersโ use of todayโs Internet, which is more content-centric than location-centric, involving, for example, social networking and the retrieval of aggregated data [2]. The integration of cloud computing and fog computing is one means of delivering information-centric Internet-of-Things (IoT) applications [3,4].
The fog computing paradigm has made cloud computing scalable [5]. To improve the overall performance of cloud-fog computing, fogs and the cloud must be carefully managed so that they can operate effectively and be distributed among physical resources. This work proposes a cloud-fog computing
architecture with innovative computing functions and a scheduling mechanism to support high-quality IoT applications.
Cloud-Fog Computing Architecture
Fog computing is an architecture that uses one of many collaborating end-user clients to support substantial storage, communication, control and management [6]. Fog computing can be perceived in both large cloud systems and large data structures, owing to the growing difficulties in accessing information objectively [7,8]. Fig. 1 presents cloud-fog computing architecture, which results in a lack of quality of obtained content. The effects of fog computing on cloud and big data systems may vary; accurate content distribution is commonly limited; this issue has been tackled by the creation of metrics for improving accuracy [9,10]. The cloud-fog architecture combines the advantages of cloud computing and fog computing.
Fig. 1: Cloud-fog computing architecture
Proposed Information-centric IoT Paradigm
The proposed information-centric IoT system supports IoT applications in various network domains, a resource and network management system with IoT middleware, fog
IEEE-ICASI 2017 - Meen, Prior & Lam (Eds)
computing with an agent and cloud computing. Fig. 2 shows this system. To support quality of service (QoS) in IoT applications, the close coordination and optimal operation of the above components are very important. Two functions, job classification and resource scheduling, are designed here to support QoS in various IoT applications.
Fig. 2: Information-Centric IoT Paradigm
A. Job Classification The function of job classification is to classify different IoT applications by authority, data type, data update rate and priority. These applications have various properties and are recorded in the database during the registering process. When one IoT application is triggered in the database, the management system will capture the corresponding properties for resource scheduling. Table 1 lists these different properties. Following the IoT properties, Fig.3 defines a basic rule of job classification.
Table 1: IoT application properties
Fig. 3: The basic rule for job allocation
B. Resource Scheduling In this work, resources are scheduled to support QoS-based IoT applications. The tasks of IoT applications can be served using fog or cloud computing based on the predefined scheduling mechanism. Fig. 4 presents the operating architecture.
Fig. 4: The operating architecture
The objective of resource scheduling is to support the QoS for various IoT applications. In this investigation, the objective function is defined as follows. ๐๐๐๐๐๐๐๐๐๐๐๐๐ง๐ง๐๐ ๐๐๐ถ๐ถ= (๐๐ ๐๐ร๐๐๐๐๐๐); (1) ๐๐๐๐๐๐=๐น๐น(๐ก๐ก๐๐๐ ๐ ๐๐ ๐๐) (2) where ๐๐๐ถ๐ถ and Wm are the total operational cost and the cost per operating unit, respectively. ๐๐ U ๐๐ is the operational unit of ๐ก๐ก๐๐๐ ๐ ๐๐ i. The operational units of various IoT applications can be evaluated by real applications or defined by resource managers. Fig. 5 shows such a definition. Fig. 6 presents the scheduling algorithm, based on the OC.
Fig. 5: The definition of Operation Unit
IEEE-ICASI 2017 - Meen, Prior & Lam (Eds)
Fig. 11: The process time for different IoT applications
Conclusion This work proposes novel resource scheduling for information-centric IoT applications. The mechanism applies in a cloud-fog computing environment. Based on the QoS requirements and resource limitations, computing that involves cloud, local fog and neighbor fogs can be scheduled. Performance in terms of resource utilization and service delay is analyzed herein. Simulation results reveal that the proposed scheduling mechanism outperforms computing without scheduling by reducing the cloud computing loading about 20.47%. The QoS of real-time IoT applications can be guaranteed by the proposed mechanism.
References
[1] M. Aazam and E.N. Huh, โFog Computing: The Cloud-IoT/IoE Middleware Paradigm,โ IEEE Potentials , Vol.35, No.3, pp.40-44, May-June 2016. [2] M.A. Yaqub, S.H. Ahmed, S.H. Bouk and D.K. Kim, Content-Centric Networks: An Overview, Applications, and Research Challenges, Information-Centric Networks (ICN) , Springer, 2016. [3] M. Nitti, L. Atzori and I.P. Cvijikj, โFriendship Selection in the Social Internet of Things: Challenges and Possible Strategies,โ IEEE Internet Things Journal , Vol.2, No.3, pp.240-247, July
[4] S. Sarkar and S. Misra, "Theoretical Modelling of Fog Computing: a Green Computing Paradigm to Support IoT Applications," IET Networks , Vol.5, No.2, pp.23-29, 2016. [5] K.P. Saharan and A. Kumar, โFog in Comparison to Cloud: A Survey,โ International Journal of Computer Applications , Vol.122, No.3, pp.10-12, July 2015. [6] S. Ningning, G. Chao, A. Xingshu and Z. Qiang, "Fog Computing Dynamic Load Balancing Mechanism based on Graph Partitioning," China Communications , Vol.13, No.3, pp.156-164, March 2016. [7] Fog Computing and the Internet of Things: Extend Cloud to Where the Things Are , White Paper, Cisco, pp.1-5, 2015. [8] I. Farris, R. Girau, L. Militano, M. Nitti, Atzori, A. Iera and G. Morabito, "Social Virtual Objects in the Edge Cloud," IEEE Cloud Computing , Vol.2, pp.20-28, Nov.-Dec. 2015. [9] A.D. Firas, C. Zenon and R.A. Alina, โA Review on Fog
Computing Technology,โ Proceedings of 2016 IEEE International Conference on Systems, Man, and Cybernetics , Oct. 10, 2016. [10] M.B. Xavi, M.T. Eva, G. Alejandro, B. Vitor and A. Albert, โWill It be Cloud or will It be Fog? F2C, A Novel Flagship Computing Paradigm for Highly Demanding Services,โ Proceedings of the 2016 Future Technologies Conference , pp.1129-1136, 2016.