![Related Work
• Aazam et al. In [11] :
– highlights the fast processing of the sensed data to generate instructions for the actuators and robots within
some acceptable latency as one of the challenges.
– other challenge noted was data structuring and filtering to avoid sending unnecessary data to the core and
the cloud.
• Li et al. In [12] :
– proposed a service popularity-based smart resources partitioning scheme for Fog computing-enabled IIoT to
improve delay time and fault tolerance.
– Their proposed scheme demonstrated improved performance. It was tested simulated using iFogSim a similar
simulator that has been adopted in this paper.
• de Brito et al. In [15] :
– highlights the challenge of orchestration in fog-enabled infrastructure environment and proposes an
infrastructure management and orchestration architecture for an fog computing.
– proposed architecture switches between centralized and distributed model. de Brito et al. not explore much
on the implementation of automatic event triggers.
Recently, there has been a number of promising research initiatives focused on solving issues related
to IoT and Fog computing.](https://image.slidesharecdn.com/satnac2019paperpresentation2019-09-04-190904200838/85/Optimization-of-Fog-computing-for-Industrial-IoT-applications-6-320.jpg)
![Related Work…
• Checkired et al. In [17] :
– stated that the collected and sensed data from IIoT need to be scheduled in real-time, especially for big
factories, and proposed a hierarchical Fog servers’ deployment at the network service layer across different
tiers
– this study proved that processing data closer to the edge does improve performance in IIoT applications.
• Fui et al. In [18] :
– highlighted the challenges of data processing, secure storage, efficient retrieval and dynamic data collection
in IIoT.
– argued that storing all the IIoT data raw is unwise considering that the end devices energy and storage space
is limited in these devices.
– Proposed a scheme that selectively processes data based on latency requirements, where the raw data is
pre-processed in the Fog and then the time-sensitive data are used and stored locally, while the rest is sent to
the cloud
• de Brito et al. In [20] :
– addressed the problem of distributed decision making, but this work focused on the challenge of service
hosting where Fog nodes are required to be adaptively configured to host services for sensor nodes.
– proposed the use of an online distributed algorithm, called adaptive Fog configuration (AFC), to optimize
service hosting and task admission decisions.
Recently, there has been a number of promising research initiatives focused on solving issues related
to IoT and Fog computing.](https://image.slidesharecdn.com/satnac2019paperpresentation2019-09-04-190904200838/85/Optimization-of-Fog-computing-for-Industrial-IoT-applications-7-320.jpg)
![References
[1] IDC, Internet of Things Spending Guide by Vertical Market, 2014.
[2] Luxiu Yin, Juan Luo, Haibo Luo. “Tasks Scheduling and Resource Allocation in Fog Computing Based on Containers for Smart Manufacturing,” IEEE
Transactions on Industrial Informatics, vol. 14, no. 10, October, 2018
[3] Microsoft, Competing in the digital age of manufacturing, 2018.
[4] Sebastian Trinks, Carsten Felden. “Edge Computing architecture to support Real Time Analytic applications: A State-of-the-art within the application area
of Smart Factory and Industry 4.0,” IEEE International Conference on Big Data (Big Data), 2018
[5] Flavio Bonomi, Rodolfo Milito, Jiang Zhu, Sateesh Addepalli. “Fog Computing and Its Role in the Internet of Things,” ACM SIGCOMM Mobile Cloud
Computing (MCC), 2012
[6] P1934/D2.0, Apr 2018 - IEEE Approved Draft Standard for Adoption of OpenFog Reference Architecture for Fog Computing, 2018
[7] Yuxuan Jiang, Zhe Huang, Danny H. K. Tsang. “Challenges and Solutions in Fog Computing Orchestration,” IEEE Network, volume: 32, issue: 3, May/June,
2018
[8] Mithun Mukherjee, Lei Shu, Di Wang. “Survey of Fog Computing: Fundamental, Network Applications, and Research Challenges,” IEEE Communications
Surveys & Tutorials, vol. 20, no. 3, 2018
[9] B. Zhang, N. Mor , J. Kolb , D.S. Chan , K. Lutz , E. Allman , J. Wawrzynek , E.A. Lee , J. Kubiatowicz , The cloud is not enough: saving IoT from the cloud.,
HotStorage, 2015
[10] Saad Khan, Simon Parkinson, Yongrui Qin. “Fog computing security: a review of current applications and security solutions,” Springer Journal of Cloud
Computing, December 2017
[11] Mohammad Aazam, Sherali Zeadally, and Khaled A. Harras. “Deploying Fog Computing in Industrial Internet of Things and Industry 4.0,” IEEE
Transactions on Industrial Informatics, vol. 14, no. 10, October 2018
[12] Gaolei Li, Jun Wu, Jianhua Li, Kuan Wang, and Tianpeng Ye. “Service Popularity-Based Smart Resources Partitioning for Fog Computing-Enabled Industrial
Internet of Things,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018
[13] Sabelo Dlamini, Neco Ventura. “Enhancement of Fog Computing based Autonomous Management Systems,” Southern Africa Telecommunication
Networks and Applications Conference (SATNAC), 2018
[14] Sabelo Dlamini, Joyce Mwangama, Neco Ventura and Thomas Magedanz. “Design of an Autonomous Management and Orchestration for Fog
Computing,” International Conference on Intelligent and Innovative Computing Applications (ICONIC), 2018
[15] M.S. de Brito, S. Hoque, R. Steinke, A. Willner and T. Magedanz. “Application of the Fog computing paradigm to Smart Factories and cyber-physical
systems,” Trans Emerging Tel Tech. 2018;](https://image.slidesharecdn.com/satnac2019paperpresentation2019-09-04-190904200838/85/Optimization-of-Fog-computing-for-Industrial-IoT-applications-15-320.jpg)
![References…
[16] Harshit Gupta, Amir Vahid Dastjerdi, Soumya K. Ghosh, Rajkumar Buyya. " iFogSim: A Toolkit for Modeling and Simulation of Resource Management
Techniques in Internet of Things, Edge and Fog Computing Environments"
[17] Djabir Abdeldjalil Chekired, Lyes Khoukhi and Hussein T. Mouftah. “Industrial IoT Data Scheduling Based on Hierarchical Fog Computing: A Key for
Enabling Smart Factory,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018
[18] Jun-Song Fu, Yun Liu, Han-Chieh Chao, Bharat K. Bhargava and Zhen-Jiang Zhang. “Secure Data Storage and Searching for Industrial IoT by Integrating Fog
Computing and Cloud Computing,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018
[19] Mohammad Ashjaei and Marcus Bengtsson “Enhancing Smart Maintenance Management Using Fog Computing Technology,” IEEE IEEM, 2017
[20] Lixing Chen, Pan Zhou, Liang Gao and Jie Xu. “Adaptive Fog Configuration for the Industrial Internet of Things,” IEEE Transactions on Industrial
Informatics, vol. 14, no. 10, October 2018
[21] Chenhua Shi, Zhiyuan Ren, Kun Yangy, Chen Chen, Hailin Zhang, Yao Xiao and Xiangwang Hou. “Ultra-Low Latency Cloud-Fog Computing for Industrial
Internet of Things,” IEEE Wireless Communications and Networking Conference (WCNC), 2018
[22] Gabriele Castellano, Fulvio Risso, and Riccardo Loti. “Fog Computing over Challenged Networks: a Real Case Evaluation,” IEEE 7th International
Conference on Cloud Networking (CloudNet), October, 2018
[23] Olena Skarlat, Vasileios Karagiannis, Thomas Rausch, Kevin Bachmann and Stefan Schulte. “A Framework for Optimization, Service Placement, and
Runtime Operation in the Fog,” IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC), 2018.
[24] F. Bonomi, R. Milito, P. Natarajan, J. Zhu, "Fog Computing: A Platform for Internet of Things and Analytics", Big Data and Internet of Things: A Roadmap
for Smart Environments ser. Studies in Computational Intelligence, vol. 546, pp. 169-186, 2014.
[25] A.V. Dastjerdi, H. Gupta, R.N. Calheiros, S.K. Ghosh, R. Buyya, "Fog Computing: Principles Architectures and Applications", Internet of Things: Principles
and Paradigms, 2016.
[26] Sergej Svorobej, Patricia Takako Endo, Malika Bendechache, Christos Filelis-Papadopoulos, Konstantinos M. Giannoutakis, George A. Gravvanis, Dimitrios
Tzovaras, James Byrne and Theo Lynn. “Simulating Fog and Edge Computing Scenarios: An Overview and Research Challenges,” MDPI Future Internet, 2018
[27] Qayyum, T.; Malik, A.W.; Khattak, M.A.K.; Khalid, O.; Khan, S.U. FogNetSim++: A Toolkit for Modeling and Simulation of Distributed Fog Environment. IEEE
Access 2018
[28] Saiful Hoque, Mathias Santos de Brito, Alexander Willner. “Towards Container Orchestration in Fog Computing Infrastructures,” International Workshop
on Fog Computing and Networking (FCN), July, 2017
[29] Zhenyu Wen, Renyu Yang, Peter Garraghan, Tao Lin, Jie Xu, and Michael Rovatsos. Fog orchestration for internet of things services. IEEE Internet
Computing, 21(2):16–24, 2017.](https://image.slidesharecdn.com/satnac2019paperpresentation2019-09-04-190904200838/85/Optimization-of-Fog-computing-for-Industrial-IoT-applications-16-320.jpg)
Optimization of Fog computing for Industrial IoT applications
- 1. The Changing Face of Telcos in a Digital World Optimisation of Fog computing for Industrial IoT applications Sabelo Dlamini University of Cape Town Cape Town, South Africa dlmsab003@myuct.ac.za Neco Ventura University of Cape Town Cape Town, South Africa neco@crg.ee.uct.ac.za H. Anthony Chan Huawei Technologies 1700 Alma Dr. Plano, USA h.anthony.chan@huawei.com 2019 Southern Africa Telecommunication Networks and Applications Conference (SATNAC) SATNAC 2019 (September 1-4, 2019) Fairmont Zimbali, Ballito, Durban, South Africa
- 2. Contents • Introduction • Background • Related Work • Proposed Scheme • Evaluation • Conclusion and Future Work
- 3. Introduction • The application of IoT in the manufacturing industry is normally referred to as the Industrial Internet of Things (IIoT). • Manufacturing companies that have adopted smart manufacturing in their operations are already seeing improved results with 82% increased operational efficiency and improved product quality with 49% fewer product defects, according to Microsoft research . • American Society for Quality states that connecting devices and analyzing the data collected enables manufacturers to reduce overhead, conserve resources, increase profits and optimize operational efficiencies Internet of Things (IoT) is expected to provide a quantum leap for the manufacturing industry with the market potential for industrial applications projected to be $913 billion by 2018. Industry analysts are projecting that there will be up to 50 billion connected devices with a market potential of between $10.6 to 14.2 trillion by 2030
- 4. Fog Computing • As more factory processes adopt smart manufacturing, the data velocity and volume increases, moving the big data from factory sensor/actuators to the cloud might not be efficient, or even feasible due to bandwidth and network connection constraints. • While on the other hand, smart manufacturing is required to support time-sensitive and location-aware applications, such as real-time manufacturing applications, making the distant cloud not be able to satisfy the low latency requirements of these applications. • Fog computing bridges the gap between the cloud and things to address the latency, network bandwidth and availability and privacy/security issues in Internet of Things (IoT). • Fog architectures selectively move compute, storage, communication, control, and decision making closer to the network edge where data is being generated in order solve the limitations in current infrastructure to enable mission-critical, data-dense use cases. Current “cloud-only” architectures cannot keep up with the volume and velocity of this data across the network, thereby reducing the value that can be created and captured from these investments.
- 5. Fog Computing – Autonomy • As a technological paradigm, edge computing may be architecturally organized as an autonomic (self-healing) computing network. • Autonomy should enable decision making to be made at all levels of a deployment’s hierarchy including near the device or higher order layers. • Autonomy at the network edge means intelligence from local devices and peer data can be used to fulfil the business’ mission at the point where it makes the most business sense. • In this study we focus on enhancing Fog Computing’s autonomy to reduce delays in decision-making. The autonomy pillar requires fog nodes to continue to deliver the designed functionality in the face of the external service failures.
- 6. Related Work • Aazam et al. In [11] : – highlights the fast processing of the sensed data to generate instructions for the actuators and robots within some acceptable latency as one of the challenges. – other challenge noted was data structuring and filtering to avoid sending unnecessary data to the core and the cloud. • Li et al. In [12] : – proposed a service popularity-based smart resources partitioning scheme for Fog computing-enabled IIoT to improve delay time and fault tolerance. – Their proposed scheme demonstrated improved performance. It was tested simulated using iFogSim a similar simulator that has been adopted in this paper. • de Brito et al. In [15] : – highlights the challenge of orchestration in fog-enabled infrastructure environment and proposes an infrastructure management and orchestration architecture for an fog computing. – proposed architecture switches between centralized and distributed model. de Brito et al. not explore much on the implementation of automatic event triggers. Recently, there has been a number of promising research initiatives focused on solving issues related to IoT and Fog computing.
- 7. Related Work… • Checkired et al. In [17] : – stated that the collected and sensed data from IIoT need to be scheduled in real-time, especially for big factories, and proposed a hierarchical Fog servers’ deployment at the network service layer across different tiers – this study proved that processing data closer to the edge does improve performance in IIoT applications. • Fui et al. In [18] : – highlighted the challenges of data processing, secure storage, efficient retrieval and dynamic data collection in IIoT. – argued that storing all the IIoT data raw is unwise considering that the end devices energy and storage space is limited in these devices. – Proposed a scheme that selectively processes data based on latency requirements, where the raw data is pre-processed in the Fog and then the time-sensitive data are used and stored locally, while the rest is sent to the cloud • de Brito et al. In [20] : – addressed the problem of distributed decision making, but this work focused on the challenge of service hosting where Fog nodes are required to be adaptively configured to host services for sensor nodes. – proposed the use of an online distributed algorithm, called adaptive Fog configuration (AFC), to optimize service hosting and task admission decisions. Recently, there has been a number of promising research initiatives focused on solving issues related to IoT and Fog computing.
- 8. Fog Computing – Machine Learning • Robustness of machine learning algorithms is a key challenge for applications (e.g., self-driving cars and network operation) in real-world environments where learning errors could lead to high costs. • Most machine learning implementations require training process which is often time consuming and involves high cost for small to medium manufactures • One big challenge faced by cloud-based ML-solutions in manufacturing is delay. Most applications are delay-sensitive, it is not easy to design a real-time system with heavy computation loads. • Challenges with cloud-based Machine Learning solutions: – implementations require training process which is often time consuming and involves high cost. – require data collection and data analysis for the performance of the network Machine Learning is in the forefront of research today that requires computation for both training models, and inference or scoring those models for close to real time response at the edge.
- 9. Fog Computing – Hidden Markov Model (ML) • Finite State Machines are efficient implementations of artificial intelligence with less storage and processing requirements • Finite State Machines are considered at the design phase and have been successfully implemented in: – Border Gateway Protocols (BGP) – Controller Area Network (CAN) – IEEE 802.11 (Wi-Fi) • Edge Intelligence: – Propose an option to provide decent intelligence to a network without going through the different stages required to setup a machine learning function to support network decision making. • OpenFog, ETSI NFV MANO and FSM • Requires detailed design and analysis of the proposed operation environment Proposed scheme uses Hidden Markov Models, a machine learning algorithm that has been widely used in autonomous robotics to enhance decision-making. ETSI MANO Architecture OpenFog Architecture Hidden Markov Model Io T Fog Manageme nt State
- 10. Proposed Scheme Integration and the use of Hidden Markov Model to automatically optimise the Fog computing based systems. • Propose a scheme to enable fog-based systems to self-manage, including self-configuration, self-optimization, and self-healing. This can be achieved by an autonomic function having minimal dependencies on human administrators or centralized management systems • Proposed scheme will be sitting in the edge node of the autonomous edge network • If the network performance indicators are not satisfying the network user requirements as defined, the proposed scheme will automatically change the state of the network • Proposed scheme will use reduce latency (average loop latency), and network usage. Backhaul Edge Node Central Cloud Autonomous Management and Orchestration Centralised ManagementDistributed Management Fog Nodes ETSI MANO HM M
- 11. Proposed Scheme As mentioned in the proposed scheme, we will adopt the Hidden Markov Model as a Machine Learning algorithm which has widely adopted in resource management in cloud computing applications . • Distributed Mode: In this state, the Fog computing is fully autonomous and is able to independently make decisions and function on its own, with sufficient resources available. In this state, the computing functions are locally implemented, the main aim is to have the system spending most of its time in this mode or state. • Hybrid Mode: This is the state where Fog computing has unreliable connectivity to the cloud. Fog computing is integrated with the cloud over the unguaranteed network connecting, necessitating switching between the previous state and the centralized state. In this state, Fog computing performs storage and computation are locally implemented while networking is remote. • Centralized Mode: In this state, the Fog is integrated with the cloud well. Computation, storage, and networking are centrally provided by the Cloud. Cloud provides more sturdy computation and storage facilities with the drawback of high backhaul link consumption and cost related to the use of cloud computing facilities.
- 12. Experimental Setup • In order to compare the performance of both the resource management scheme for the study, we used iFogSim was used for simulation of resource management, specifically scheduling policies between Fog computing and cloud computing with varying network configurations . • We used a basic Fog node to sensor setup, each sensor/actuator is connected to a Wi-Fi gateway, the gateway is then connected to an ISP gateway, which connects the Fog computing to the Cloud datacenter. . Source Destination Latency (ms) Sensor/Actuator Fog Node 61ms Fog Node Fog Node 22ms Fog Node Fog Node 70ms Edge Node Edge Node 30ms Edge Node Cloud 200ms
- 13. • Distributed management provide very low latencies for Fog computing with a limited number of sensor/actuators connected to the Fog node. • distributed management provides lower latency when compared to centralized management in nodes below 16. • Network usage from 16 sensors/actuators in both centralized and distributed mode performance is the same, this is caused by limited resources in the Fog nodes and transferred processing to the cloud. • This results in similar performance between the two-management mode. Therefore, it may be better to switch to centralized mode as soon as the network usage is similar. Results and Discussion
- 14. Conclusion • Distributed management mode performed better than centralized with a low number of sensor/actuators connected to the Fog node. • Both management modes demonstrated similar performance from 32 sensors/actuators, highlighting the need for both modes. • One big challenge faced by cloud-based ML-solutions in manufacturing is delay. Most applications are delay-sensitive, it is not easy to design a real-time system with heavy computation loads. • The results highlight that there is a need for both orchestration (centralized management) and choreography (distributed management) in a Fog computing environment. • In addition to this we believe that there is a need for an intelligent switching between the centralized and distributed mode in order to enhance performance and reduce latency. As future work to develop an intelligent machine learning based scheme that can enhance switching between the two modes. The scheme will sense the change conditions and switch management mode accordingly.
- 15. References [1] IDC, Internet of Things Spending Guide by Vertical Market, 2014. [2] Luxiu Yin, Juan Luo, Haibo Luo. “Tasks Scheduling and Resource Allocation in Fog Computing Based on Containers for Smart Manufacturing,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October, 2018 [3] Microsoft, Competing in the digital age of manufacturing, 2018. [4] Sebastian Trinks, Carsten Felden. “Edge Computing architecture to support Real Time Analytic applications: A State-of-the-art within the application area of Smart Factory and Industry 4.0,” IEEE International Conference on Big Data (Big Data), 2018 [5] Flavio Bonomi, Rodolfo Milito, Jiang Zhu, Sateesh Addepalli. “Fog Computing and Its Role in the Internet of Things,” ACM SIGCOMM Mobile Cloud Computing (MCC), 2012 [6] P1934/D2.0, Apr 2018 - IEEE Approved Draft Standard for Adoption of OpenFog Reference Architecture for Fog Computing, 2018 [7] Yuxuan Jiang, Zhe Huang, Danny H. K. Tsang. “Challenges and Solutions in Fog Computing Orchestration,” IEEE Network, volume: 32, issue: 3, May/June, 2018 [8] Mithun Mukherjee, Lei Shu, Di Wang. “Survey of Fog Computing: Fundamental, Network Applications, and Research Challenges,” IEEE Communications Surveys & Tutorials, vol. 20, no. 3, 2018 [9] B. Zhang, N. Mor , J. Kolb , D.S. Chan , K. Lutz , E. Allman , J. Wawrzynek , E.A. Lee , J. Kubiatowicz , The cloud is not enough: saving IoT from the cloud., HotStorage, 2015 [10] Saad Khan, Simon Parkinson, Yongrui Qin. “Fog computing security: a review of current applications and security solutions,” Springer Journal of Cloud Computing, December 2017 [11] Mohammad Aazam, Sherali Zeadally, and Khaled A. Harras. “Deploying Fog Computing in Industrial Internet of Things and Industry 4.0,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018 [12] Gaolei Li, Jun Wu, Jianhua Li, Kuan Wang, and Tianpeng Ye. “Service Popularity-Based Smart Resources Partitioning for Fog Computing-Enabled Industrial Internet of Things,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018 [13] Sabelo Dlamini, Neco Ventura. “Enhancement of Fog Computing based Autonomous Management Systems,” Southern Africa Telecommunication Networks and Applications Conference (SATNAC), 2018 [14] Sabelo Dlamini, Joyce Mwangama, Neco Ventura and Thomas Magedanz. “Design of an Autonomous Management and Orchestration for Fog Computing,” International Conference on Intelligent and Innovative Computing Applications (ICONIC), 2018 [15] M.S. de Brito, S. Hoque, R. Steinke, A. Willner and T. Magedanz. “Application of the Fog computing paradigm to Smart Factories and cyber-physical systems,” Trans Emerging Tel Tech. 2018;
- 16. References… [16] Harshit Gupta, Amir Vahid Dastjerdi, Soumya K. Ghosh, Rajkumar Buyya. " iFogSim: A Toolkit for Modeling and Simulation of Resource Management Techniques in Internet of Things, Edge and Fog Computing Environments" [17] Djabir Abdeldjalil Chekired, Lyes Khoukhi and Hussein T. Mouftah. “Industrial IoT Data Scheduling Based on Hierarchical Fog Computing: A Key for Enabling Smart Factory,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018 [18] Jun-Song Fu, Yun Liu, Han-Chieh Chao, Bharat K. Bhargava and Zhen-Jiang Zhang. “Secure Data Storage and Searching for Industrial IoT by Integrating Fog Computing and Cloud Computing,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018 [19] Mohammad Ashjaei and Marcus Bengtsson “Enhancing Smart Maintenance Management Using Fog Computing Technology,” IEEE IEEM, 2017 [20] Lixing Chen, Pan Zhou, Liang Gao and Jie Xu. “Adaptive Fog Configuration for the Industrial Internet of Things,” IEEE Transactions on Industrial Informatics, vol. 14, no. 10, October 2018 [21] Chenhua Shi, Zhiyuan Ren, Kun Yangy, Chen Chen, Hailin Zhang, Yao Xiao and Xiangwang Hou. “Ultra-Low Latency Cloud-Fog Computing for Industrial Internet of Things,” IEEE Wireless Communications and Networking Conference (WCNC), 2018 [22] Gabriele Castellano, Fulvio Risso, and Riccardo Loti. “Fog Computing over Challenged Networks: a Real Case Evaluation,” IEEE 7th International Conference on Cloud Networking (CloudNet), October, 2018 [23] Olena Skarlat, Vasileios Karagiannis, Thomas Rausch, Kevin Bachmann and Stefan Schulte. “A Framework for Optimization, Service Placement, and Runtime Operation in the Fog,” IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC), 2018. [24] F. Bonomi, R. Milito, P. Natarajan, J. Zhu, "Fog Computing: A Platform for Internet of Things and Analytics", Big Data and Internet of Things: A Roadmap for Smart Environments ser. Studies in Computational Intelligence, vol. 546, pp. 169-186, 2014. [25] A.V. Dastjerdi, H. Gupta, R.N. Calheiros, S.K. Ghosh, R. Buyya, "Fog Computing: Principles Architectures and Applications", Internet of Things: Principles and Paradigms, 2016. [26] Sergej Svorobej, Patricia Takako Endo, Malika Bendechache, Christos Filelis-Papadopoulos, Konstantinos M. Giannoutakis, George A. Gravvanis, Dimitrios Tzovaras, James Byrne and Theo Lynn. “Simulating Fog and Edge Computing Scenarios: An Overview and Research Challenges,” MDPI Future Internet, 2018 [27] Qayyum, T.; Malik, A.W.; Khattak, M.A.K.; Khalid, O.; Khan, S.U. FogNetSim++: A Toolkit for Modeling and Simulation of Distributed Fog Environment. IEEE Access 2018 [28] Saiful Hoque, Mathias Santos de Brito, Alexander Willner. “Towards Container Orchestration in Fog Computing Infrastructures,” International Workshop on Fog Computing and Networking (FCN), July, 2017 [29] Zhenyu Wen, Renyu Yang, Peter Garraghan, Tao Lin, Jie Xu, and Michael Rovatsos. Fog orchestration for internet of things services. IEEE Internet Computing, 21(2):16–24, 2017.
- 17. DANKIE