The challenge

• The main idea of 3SCity-E2C Workshop:

The main idea of “building software services in smart city through edge-to-cloud orchestration” was proposed in the first series of the 3SCity-E2C workshop. This workshop series focuses on how the “resource management mechanism” through multilevel information communication technology (ICT) architecture can be beneficial for the effective and intelligent software service execution in large-scale IoT networks of smart cities.

• The main idea of the “3rd 3SCity-E2C Workshop:”

This series of our workshops aim to integrate all idea of the past “3SCity-E2C” workshop, including “3SCity-E2C,” “D2C-ML&AI,” “Special Track of D2C-ML&AI.” So, we propose the idea of how we can build the “intelligent resource management mechanism for Service Execution of large-scale IoT Networks.” This highlights the collaboration of “Software Engineering Development,” “ML/AI applied to IoT,” “facilities of multilevel ICT architectures,” and “edge-to-cloud computing orchestration.”

1. Research Questions and Problem Statement:

Cloud services and solutions are hot resolutions for small businesses to global enterprises. However, they remain a broad concept covering much online territory. As you start to shift your business to cloud solutions, whether, for application or infrastructure deployment, it is more critical than ever to know the differences and benefits of multiple cloud services. Though “as-a-service” types are increasing by the day, there are regularly three models of cloud service to compare, including Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS), as shown in Fig 1.

Figure 1.  SaaS, PaaS, and IaaS [1]

Recently, several questions have come up about how we can adopt three models of cloud service to edge computing. Some works for each of those are available to present how it could be done at Fog/Edge layers, but there are still several doubts and questions. So, one question is an adaptation to edge layers. Another question is how we can build an intelligent orchestration from edge-to-cloud that makes it possible to connect all three models of cloud service solutions to edge solutions. We called it “E2CaaS: Edge-to-Cloud-as-a-Service” and their related solutions to all three cloud service models.

2. Objectives:

As shown in Fig 2., multilevel ICT architectures are designed based on different ICT and business domain requirements. Recently, there are several efforts to integrate “software service development,” “applied advanced machine learning (ML) and artificial intelligence (AI) to IoT,” and “ICT management requirements” in large-scale IoT networks of smart cities.

In this series of workshops, we are interested in discussing and designing integral ICT architecture solutions for “software service development” through “advanced ML and AI” for smart cities’ large-scale IoT networks. So, the main discussion points and targets are shown below:

• We go beyond different “advance ML and AI techniques” that may be applied to large-scale ICT management of smart cities, such as “AI at Fog/Edge,” “AI at Cloud,” or any other integral AI solutions for large-scale IoT networks of smart cities which are yet demanding and we are looking for further scientific discussion.
• We present different software service development solutions in smart cities, such as “Edge-as-a-Service,” “MEC-as-a-Service,” “Fog-as-a-Service, “cloudlet-as-a-Service, etc. We have recently proposed our integral architectural solution to develop software services in smart cities’ large-scale IoT networks based on edge-to-cloud orchestration, Edge-to-Cloud-as-a-Service(E2CaaS).
• We are going to discuss how “software service development” and “advance ML and AI techniques” can mix in the smart cities to provide facilities for citizens’ demands, such as “intelligent resource management techniques for the service execution.” In addition, we also care about the General Data Protection Regulation (GDPR), data privacy, and the security concerns of citizens in our ICT architecture design.

To sum up, we are looking to design an integral ICT architectural solution for “ML&AI-E2C: ML&AI at Edge-to-Cloud orchestration.” One of the facilities of “ML&AI-E2C” solutions can be realized in the “intelligent resource management for the service execution” at the middleware platform of large-scale IoT networks of smart cities.

Figure 2.  Our proposed hybrid ICT architecture in the smart city based on different city-data-source formats [2] and some relevant information about the main “large-scale ICT architecture” idea available here

3. Challenges (The main reference of this part is here):

Why is it necessary to consider and extend “Cloud Services” and “ML and AI techniques” at the edge of IoT networks?

1. Data Growth in terms of “Size,” “Time,” and “Scale” (Large-Scale ICT and its Data Management): Exponential growth of city-data in widely distributed storage media of Smart Cities, as data is the main ingredient for ML and AI techniques and algorithms;
2. Data Privacy: Citizens and data stakeholders may not be willing to share their information in a public data storage (e.g., Cloud technologies platform);
3. General Data Protection Regulation (GDPR): GDPR is a regulation in the EU law on data protection and privacy. GDPR also addresses the transfer of personal data outside the EU and EEA areas;
4. Cybersecurity Concerns: In case of an attack at the centralized platform place (Cloud technologies) or multi-attacks at the IoT devices network will occur, how can we collect and send data/datasets to a centralized platform place to extract our knowledge requirements through data analysis/analytic and ML and AI techniques?
5. Complexities of running and managing complex ML and AI techniques at the edge of networks: Due to computational and memory limitations of IoT devices and low processing abilities at the edge of the Smart City networks, IoT devices are often not capable of running and managing complex ML and AI techniques and algorithms at the edge of the Smart City networks.
6. Cost of Cloud services: In most cases, having Cloud storage and services can be quite expensive. Being charged year after year for a monthly or annual subscription can add up. Cloud storage and services providers know that customers will pay those high fees to have their data backed up and ultimately to have peace of mind.

4. Introduction:

A large-scale number of heterogeneous internet of things (IoT) devices engage in any smart city’s IoT networks. To organize and provide effective integration between all IoT data sources in smart city IoT networks, decentralized-to-centralized information communication technology (DC2C-ICT) and distributed-to-centralized ICT (D2C-ICT) architectures have been designed as a collaborative, coordinated, and hierarchical edge-to-cloud computing orchestration platform. Our recent studies present how our proposed ICT architectures may map to requirements of large-scale IoT networks of smart cities and provide facilities for managing data, resources, software services, and cybersecurity and network [3].

Developing software services in large-scale IoT networks of smart cities is a quite challenging job in different multilevel ICT architecture, as shown in Fig. 3. It is also required to execute some “software services execution tasks.” Usually, each task has diverse necessities, which have to be fulfilled. One of the main “software service execution tasks” is related to large-scale ICT “resource management.” Recently, there are several efforts to design the intelligent mechanism for “software service execution tasks” and their related “resource management” mechanism in large-scale IoT networks of smart cities. However, there are still infant design stage and more technical decision and research required. One of the great intelligent design solutions proposes through edge-to-cloud computing platform orchestration and integration to the usage of advanced machine learning (ML) and artificial intelligence (AI) techniques to create an intelligent system for “software service execution” tasks and their “resource management” mechanism.

Figure 3. Different Software Service layers from edge-to-cloud orchestration [4]

“Resource statistical information” presents a significant role in efficiently implementing the “resource management” mechanism for the consumers’ software services execution platform in smart cities. On the one hand, continuous monitoring of engaging devices can produce many “resource statistical information” through large-scale IoT networks of smart cities. On the other hand, with this statistical information’s help, it is getting much simpler to realize the device’s availability and suitability for executing some software services tasks for building software services in smart cities. Accordingly, to ensure more reliable software service facilities for any latency-sensitive software services requirements in the smart cities, designing an intelligent ICT architecture and their computing platform eco-system is quite challenging and necessary. This ICT architecture may make it possible to organize securely distribute the city-data and resource statistical information over smart cities’ large-scale IoT networks.

Designing an advanced and more intelligent ICT eco-system is required to consider effective mechanism management to utilize IoT resources in smart cities’ large-scale IoT networks. Notably, effective utilization of resources guarantees decreasing unnecessary resource consumption and develops the overall ICT eco-system performance. Generally, the utilization and resource handling process are essentially related to the “resource selection and allocation mechanism” strategies. However, on-demand “resource selection and allocation” is a complex task for any large-scale D2C-ICT architecture eco-system. Also, the prediction of ICT resource usage (such as network communication bandwidth, CPU, hard disk, and ram) and their related performance (e.g., task execution time) can help identify their appropriateness executing some tasks for building software services in smart cities. Therefore, adopting the advanced ML and AI techniques can be valuable for designing an advanced and modern resource allocation mechanism in smart cities’ large-scale IoT networks. However, adopting and implementing advanced ML and AI techniques in a large-scale ICT architecture eco-system in smart cities can be quite complex and in an infant design and development stage among researchers in academia and industries. The overall diversification and many other issues can especially pose a massive challenge for successfully executing the advanced ML and AI techniques to predict the software services requirements in smart cities’ large-scale IoT networks.

In the case of the combined and hierarchical computing platform and ICT architecture (i.e., Fog-to-Cloud or Fog-to-cloudlet-to-Cloud) from the distributed-to-centralized schema, it is essential to define the architectural framework for predicting the “resource management” requirements based on various ML and AI techniques. This framework makes numerous facilities to analyze varieties of “Resource statistical information” datasets from various IoT devices in large-scale IoT networks of smart cities. Therefore, it is essential to define and design that kind of architectural framework for combined Edge to Cloud technologies systems to manage and predict resource management’ requirements, demanding a specific effort from the research community. Currently, “federated learning” and “replicated learning” techniques provide attractive facilities where can be applied through different architectural layers of Edge-to-Cloud orchestration to train varieties of resource statistical datasets as well as using different ML and AI techniques. One example of the combination of DC2C-ICT architecture and advanced FL models is shown in Fig. 4.

Figure 4. Federated Learning for prediction of “resource management” through DC2C-ICT architecture [5]

References:

[1] Figure extracted from “https://azure.microsoft.com/en-us/overview/what-is-iaas/.”

[2] A. Sinaeepourfard, et al., “ICT architectures for large-scale IoT networks of smart cities,” preparing for the journal submission – please follow the new update regularly.

[3] A. Sinaeepourfard, J. Krogstie, S. Sengupta, “Distributed-to-Centralized Data Management: A New Sense of Large-Scale ICT Management of Smart City IoT Networks,” IEEE Internet of Things Magazine (IEEE IoTM), 2020.

[4] J. Robberechts, A. Sinaeepourfard,  T. Goethals, and B. Volckaert, “A Novel Edge-to-Cloud-as-a-Service (E2CaaS) Model for Building Software Services in Smart Cities,” IEEE Int’l. Conf. Mobile Data Management (MDM), 2020.

[5] S. Sengupta, J. Garcia, X. Masip-Bruin,  “Collaborative learning-based Schema for Predicting Resource Usage and Performance in F2C Paradigm,” IEEE International Conference on Advanced Networks and Telecommunications Systems (IEEE ANTS), 2020

Organization

Submission

Program

TBA

Venue

Please visit this link for further information.

Contact

For further information, please contact us at a.sinaee@ntnu.no. Please put the title of your email “3SCity-E2C 2021 Workshop-Question”.