Space Information Network Resource Scheduling for Cloud Computing: A Deep Reinforcement Learning Approach

Abstract

With the development of satellite technology, space information networks (SINs) have been applied to various fields. SINs can provide more and more complex services and receive more and more tasks. The existing resource scheduling algorithms are difficult to play an efficient role in such a complex environment of resources and tasks. We propose a resource allocation scheme based on reinforcement learning. Firstly, according to the characteristics of the resources of SINs, we established the cloud computing architecture of SINs to manage the resources uniformly. Next, we adopt a variable granularity resources clustering algorithm based on fuzzy and hierarchical clustering algorithms. This algorithm can adaptively adjust the resource size and reduce the scheduling range. Finally, we model the resource scheduling process by a reinforcement learning algorithm to solve the joint resource scheduling problem. The simulation results show that the scheme can effectively reduce resources consumption, shorten the task execution time, and improve the resource utilization efficiency of SINs.

A Dual-Agent Approach for Coordinated Task Offloading and Resource Allocation in MEC

“…(10) Store the (ob 1 (t), ac 1 (t), re 1,2 (t), ob 1 (t + 1)) in the replay bufer of agent 1. (11) Store the (ob 2 (t), ac 2 (t), re 1,2 (t), ob 2 (t + 1)) in the replay bufer of agent 2. (12) if batch size < the current capacity of bufer Ten (13) for agent i � 1, 2 do (14) Sample a batch of experiences randomly.…”

“…When determining which tasks to ofoad to edge nodes, it is crucial to consider the allocation of communication bandwidth and edge node computing resources. In MEC, resource allocation needs to meet the task requirements while also taking into account the limitations of computing capability and network bandwidth of the edge nodes [9][10][11].…”

Incentive-Vacation Queueing in Extreme Edge Computing: An Analytical Reward-Based Framework

“…The authors of [24] overcome selfish behaviour by having the RSU lease resources from multiple vehicles, while having the vehicles select contracts that maximize their rewards. Similarly, the authors in [25] study the usage of electric vehicles as computational resource nodes during their charging time. They study the aspect of scheduling computational tasks in exchange for energy with the purpose of maximizing social welfare.…”

Task‐oriented resource prediction and adjustment algorithm for space‐based information network

“…The distance of information transmission in the space‐based information network is much larger than on the ground. The nodes are highly dynamic and have large topological space‐time scales, which makes the overhead incurred in performing a task in the space‐based environment much higher than that on the ground 7–9 . Second, because the resources of the space‐based environment are minimal, it is necessary to match the task demand with the resource supply efficiently.…”