Skip to main content

AI-empowered Resource Provision and Service Scheduling in Multi-Clouds

With the rapid development of the Internet of Things (IoT) technologies and the increasing popularity of IoT devices, more and more computation-intensive IoT applications become available. However, due to the limited resources of IoT devices, it is very hard for IoT devices to process the computation-intensive applications by themselves locally. Cloud computing is a popular solution to this issue, which can provide resources in a cost effective and elastic way. IoT users do not need to build or maintain the cloud computing system. Instead, they just need to rent resources from the cloud providers. However, it is hard to rely solely on a single cloud to provide all the resources and schedule all the services for the IoT users. First, a single cloud may not have sufficient resources to offer services for all the IoT applications from geographically different regions, and the cloud sometimes needs to fix the system vulnerabilities, which will result in a serious performance degradation if there is only one cloud offering service. Second, it is demonstrated that with the increasing number in cloud providers, IoT users are willing to be served by different clouds to avoid the cloud provider lock-in. Therefore, it is more and more popular and promising to serve IoT users by multi-clouds.

In multi-clouds, in order to provide IoT users with good quality of experience (QoE) in a cost-effective manner, a careful and well-designed resource provision and service scheduling strategy is very critical. With the dramatic increase in the number of IoT users, a massive amount of data is generated and needs to be uploaded and processed by the clouds. But the network bandwidth of each cloud is often limited, and for the cloud with low network bandwidth, scheduling too many service requests will cause network congestion or even system crash. The resource capacity of each cloud is also different and varying, and the service request arrivals of IoT users are highly dynamic and stochastic. When the provided resources are not sufficient enough, IoT users will experience degraded QoE. However, reserving and renting too many resources may lead to a large waste of resources. Therefore, it is severely challenging to design the resource provision and service scheduling strategy in multi-clouds for IoT users. Artificial Intelligence techniques are promising solutions to address the above challenges. AI techniques have shown great potentials especially when dealing with complex, dynamic and uncertain systems. This special issue aims to attract and disseminate high-quality research results and practical solutions from both academia and industry to advance the AI-empowered multi-clouds for IoT. The topics of interest include, but are not limited to:

  • AI for pricing in resource provision in multi-clouds for IoT
  • AI for QoE-aware service scheduling in multi-clouds for IoT
  • AI for privacy protection in multi-clouds for IoT
  • Intelligent mobility management in service scheduling for multi-clouds
  • Intelligent traffic forecasting and prediction in multi-clouds for IoT
  • Intelligent computation offloading for energy efficiency of IoT users in multi-clouds
  • Intelligent big data analytics and processing in multi-clouds for IoT
  • Testbed of AI algorithms for resource provision in multi-clouds for IoT
  • Intelligent context-aware resource provision in multi-clouds
  • Industrial applications, data and platforms in multi-clouds for IoT

Guest Editors

Ying Chen, Associate Professor, Beijing Information Science and Technology University, China;
Shangguang Wang, Professor, Beijing University of Posts and Telecommunications;
Geyong Min, Professor, University of Exeter, United Kingdom;
Qiang Ye, Assistant Professor, Memorial University of Newfoundland, Canada;
Phu Thinh Do, Assistant Professor, Post and Telecommunication Institute of Technology, Vietnam;

Provisional Deadline

Submission Deadline: 1st August 2022


Submissions should be original papers and should not be under consideration for publication elsewhere.
Extended versions of papers from relevant conferences and workshops are invited as long as the additional contribution is substantial (at least 30% of new content).
Authors should follow the formatting and submission instructions for Journal of Cloud Computing at
For more information visit the Springer Nature Information for journal Article Authors pages at
All papers will be peer-reviewed.

There are currently no articles in this collection.