Publications

2024

1. [J] TAAS’24

   [B] Joint Resource Allocation and Task Slicing for Mobile Multimedia Computing in Edge-based Autonomous Systems

   Jiwei Huang, Yajing Leng, Jiarong Bao, Songyuan Li, and Ying Chen

   ACM Transactions on Autonomous and Adaptive Systems (Under Review), 2024

   Abs

   Mobile multimedia applications such as real-time video processing, augmented reality, and mobile gaming have raised high requirements for low latency and high efficiency. Edge-based autonomous systems have become a key technology for processing these application tasks. This paper focuses on joint resource allocation and task slicing for mobile multimedia computing in edge-based autonomous systems. We propose an efficient resource allocation and task slicing strategy, aiming at the optimization of the overall utility of both edge servers and mobile devices simultaneously. We transform the resource allocation problem into resource pricing and purchasing behaviors. We present a Stackelberg game model, and prove theorems for the existence of equilibrium and optimality. Based on the theorems, we design an algorithm namely G-RPTSS for resource purchasing and computation task slicing. Then, we employ Deep Reinforcement Learning (DRL) techniques in resource pricing, and propose the DRL-ESRP algorithm which is capable of adaptively responding to dynamic computational scenarios in edge-based autonomous systems. Our scheme leverages the DRL technique for autonomous learning and policy adjustment. Simulation experiments, based on real-world scenario data, demonstrate the superior of our approach in learning efficiency and performance advantages to existing both non-DRL and other DRL algorithms.

2023

1. [J] ToN’23

   [A+] Dynamic Pricing for On-Demand DNN Inference in the Edge-AI Market

   Songyuan Li, Jia Hu, Geyong Min, Haojun Huang, and Jiwei Huang

   IEEE/ACM Transactions on Networking (Under Review), 2023

   Abs

   The convergence of edge computing and artificial intelligence (AI) gives rise to Edge-AI, which enables the deployment of real-time AI applications and services at the network edge. One of the fundamental research issues in Edge-AI is edge inference acceleration, which aims to realize low-latency high-accuracy Deep Neural Network (DNN) inference services by leveraging the fine-grained offloading of partitioned inference tasks from end devices to edge servers. However, the existing research has yet to take a more practical EdgeAI market perspective, which should consider not only the personalized inference requirements of AI users (e.g., inference accuracy, latency, and task complexity), but also the revenue incentives for AI service providers offering edge inference services. To bridge this gap, we propose an Auction-based Edge Inference Pricing Mechanism (AERIA) for revenue maximization to tackle the multi-dimensional optimization problem of DNN model partition, edge inference pricing, and resource allocation. We investigate the multi-exit device-edge synergistic inference scheme for on-demand DNN inference acceleration, and analyse the auction dynamics amongst the AI service provider, AI users and edge infrastructure provider. Owing to the strategic mechanism design via randomized consensus estimate and cost sharing techniques, the Edge-AI market attains several desirable properties, including competitiveness in revenue maximization, incentive compatibility, and envy-freeness, which are crucial to maintain the effectiveness, truthfulness, and fairness of our auction outcomes. The extensive simulation experiments based on two real-world datasets demonstrate that our AERIA mechanism significantly outperforms several state-of-the-art approaches in revenue maximization, demonstrating the efficacy of AERIA for on-demand DNN inference in the Edge-AI market.

2022

1. [J] TCCN’22

   QoE-DEER: A QoE-Aware Decentralized Resource Allocation Scheme for Edge Computing

   Songyuan Li, Jiwei Huang, Jia Hu, and Bo Cheng

   IEEE Transactions on Cognitive Communications and Networking, Jun 2022

   Abs HTML PDF

   With the increasing prevalence of online services mounted on IoT devices, edge computing gains significant momentum over conventional cloud-centric architecture. Edge servers are geographically deployed in a distributed manner nearby IoT devices, which not only frees online services from the high hardware requirement but also sharply reduces network latency experienced by IoT users. Recent works have extensively studied the problem of edge resource management and request scheduling to achieve high Quality of Service (QoS) with low latency, but there has been little focus on Quality of Experience (QoE) that an edge resource allocation scheme brings about. In this article, we study the Edge Resource Allocation (ERA) problem across multiple service requests with the objective of overall QoE maximization, which has non-polynomial computational complexity. To attack the \mathcal NP -hardness of solving the ERA problem, we adopt a game-theoretic approach to formulate the ERA problem as a potential game ERAGame which admits a Nash Equilibrium (NE). Then, we novelly present a decentralized algorithm namely QoE-DEER to find an NE solution which equivalently maximizes the overall QoE as the ERA problem. Finally, the performance and convergence of our algorithm is evaluated both theoretically and experimentally, which indicates its significant advantages over the state-of-the-art approaches.

2021

1. [J] TNSM’21

   Resource Pricing and Demand Allocation for Revenue Maximization in IaaS Clouds: A Market-Oriented Approach

   Songyuan Li, Jiwei Huang, and Bo Cheng

   IEEE Transactions on Network and Service Management, Sep 2021

   Abs HTML PDF

   With more users outsourcing their applications to the cloud, resource pricing becomes an important issue for IaaS cloud management. Jointly considering her own bidding budget and the price of cloud resources, each user is self-motivated to purchase cloud resources according to her resource demand which maximizes her own utility. Meanwhile, the cloud service provider (CSP) regulates the price of cloud resources with a certain profitability objective achieved. With an elaborate resource pricing strategy, the goals from users and the CSP are balanced and respectively satisfied to some extent. This article provides an insight into the market-oriented cloud pricing strategy. In specific, we propose an auction market in the IaaS cloud, where multiple users with heterogeneous bidding budgets and QoS requirements subscribe cloud resources according to their resource demands. The resource pricing and demand allocation scheme targeting revenue maximization also satisfies essential properties including budget feasibility, incentive compatibility and envy-freeness. To attack the NP-hardness and non-convexity of revenue maximization problem, we design a price-incentive resource auction mechanism namely RARM, which preserves an ( 1+α) approximation ratio on revenue maximization. Finally, we evaluate our RARM mechanism based on the real-world dataset to certify the efficacy of our proposed approach.

2. [J] TNSM’21

   A Price-Incentive Resource Auction Mechanism Balancing the Interests Between Users and Cloud Service Provider

   Songyuan Li, Jiwei Huang, and Bo Cheng

   IEEE Transactions on Network and Service Management, Jun 2021

   Abs HTML PDF

   For a cloud service provider, it necessitates an emerging cloud ecosystem to consolidate the existing users and attract more potential users, further gaining its market share. Therefore, in this article, we design a price-incentive resource auction mechanism in cloud environment. In response to the cloud resource price, each user synthesizes her bidding budget and QoS requirement, and purchases cloud resources according to her resource demand in a strategic manner. The cloud service provider, meanwhile, can regulate the resource demands of users through conducting a market-based pricing strategy, against too low prices to cover the operational costs (i.e., energy costs) or too high prices resulting in user churn. In virtue of an elaborate market-based pricing strategy, the interests of users and the cloud service provider are balanced. Our price-incentive resource auction mechanism targets to stimulate maximum users willing to purchase resources and perform their applications at the cloud, on the premise of a minimum profit rate guaranteed for the cloud service provider. It is also able to provide budge balance and truthfulness guarantee, and satisfy the envy-freeness. In order to carry out the above objectives, we carefully design the user utility function reflecting the complicated user interest, and formulate our resource pricing and auction problem as a bin packing problem, which has non-polynomial computational complexity. Regarding the NP-hardness of optimization problem and the concavity of user utility, we present a computational-efficient ( 1+ε)-approximate algorithm namely PIRA. Finally, we conduct simulations based on the real-world dataset to validate the effectiveness of our proposed approach.

2020

1. [J] PPNA’20

   Revenue-optimal Task Scheduling and Resource Management for IoT Batch Jobs in Mobile Edge Computing

   Jiwei Huang, Songyuan Li, and Ying Chen

   Peer-to-Peer Networking and Applications, Sep 2020

   Abs HTML PDF

   With the growing prevalence of Internet of Things (IoT) devices and technology, a burgeoning computing paradigm namely mobile edge computing (MEC) is delicately proposed and designed to accommodate the application requirements of IoT scenario. In this paper, we focus on the problems of dynamic task scheduling and resource management in MEC environment, with the specific objective of achieving the optimal revenue earned by edge service providers. While the majority of task scheduling and resource management algorithms are formulated by an integer programming (IP) problem and solved in a dispreferred NP-hard manner, we innovatively investigate the problem structure and identify a favorable property namely totally unimodular constraints. The totally unimodular property further helps to design an equivalent linear programming (LP) problem which can be efficiently and elegantly solved at polynomial computational complexity. In order to evaluate our proposed approach, we conduct simulations based on real-life IoT dataset to verify the effectiveness and efficiency of our approach.

2019

1. [C] ICWS’19

   [A] FASS: A Fairness-Aware Approach for Concurrent Service Selection with Constraints

   Songyuan Li, Jiwei Huang, Bo Cheng, Lizhen Cui, and Yuliang Shi

   In IEEE International Conference on Web Services (ICWS), Jul 2019

   Abs HTML PDF

   The increasing momentum of service-oriented architecture has led to the emergence of divergent delivered services, where service selection is meritedly required to obtain the target service fulfilling the requirements from both users and service providers. Despite many existing works have extensively handled the issue of service selection, it remains an open question in the case where requests from multiple users are performed simultaneously by a certain set of shared candidate services. Meanwhile, there exist some constraints enforced on the context of service selection, e.g. service placement location and contracts between users and service providers. In this paper, we focus on the QoS-aware service selection with constraints from a fairness aspect, with the objective of achieving max-min fairness across multiple service requests sharing candidate service sets. To be more specific, we formulate this problem as a lexicographical maximization problem, which is far from trivial to deal with practically due to its inherently multi-objective and discrete nature. A fairness-aware algorithm for concurrent service selection (FASS) is proposed, whose basic idea is to iteratively solve the single-objective subproblems by transforming them into linear programming problems. Experimental results based on real-world datasets also validate the effectiveness and practicality of our proposed approach.

2018

1. [J] IJWGS’18

   Performance Modelling and Analysis for IoT Services

   Jiwei Huang, Songyuan Li, Ying Chen, and Junliang Chen

   International Journal of Web and Grid Services, Jan 2018

   Abs HTML PDF

   With the growing popularity of Internet of Things (IoT) services on the internet, performance has become an important issue in the design and optimisation of IoT service systems. This paper proposes a theoretical approach of performance evaluation for IoT services, which is expected to provide a mathematical prediction on performance metrics at the design phase before system implementation. In specific, we first formulate an atomic service by a queueing system, and then an IoT system can be modelled by a queueing network. Detailed quantitative model analyses under different request arrival distributions are presented, and performance metrics are obtained. Furthermore, we briefly study two popular problems which are resource management and task scheduling in order to illuminate how the models and analytical results can be applied in the design and optimization of IoT systems. Finally, simulation experiments based on real-world data are conducted to validate the effectiveness of our approach.

2017

1. [C] ISPA’17

   Energy Efficient Resource Management and Task Scheduling for IoT Services in Edge Computing Paradigm

   Songyuan Li, and Jiwei Huang

   In IEEE International Symposium on Parallel and Distributed Processing with Applications (ISPA), Dec 2017

   Abs HTML PDF

   With the growing popularity of the Internet of Things (IoT), energy efficiency has been a critical concern during the design and development of IoT service systems. Meanwhile, edge computing has drawn significant attention as a burgeoning computing paradigm. This paper studies the energy efficiency issue of IoT systems by proposing a joint scheme of resource allocation and task scheduling under the edge computing paradigm. Specifically, dynamic processes of the IoT services and system are formulated by generalized queueing network models, based on which quantitative analyses of performance and energy consumption are conducted. The resource management and task scheduling are formulated by Markov Decision Process (MDP), which can balance the tradeoff between energy costs and QoS requirements. To attack the challenge of MDP search space explosion due to the large scale of IoT systems, Ordinal Optimization (OO) techniques are applied to the MDP algorithms, which are able to significantly narrow the search of MDP by slightly softening the optimization objective to a good enough subset. Finally, we conduct simulation experiments based on real-world IoT data. Evaluations and comparisons demonstrate that our approach is effective and efficient in practice.

2. [C] SCC’17

   GSPN-Based Reliability-Aware Performance Evaluation of IoT Services

   Songyuan Li, and Jiwei Huang

   In IEEE International Conference on Services Computing (SCC), Jun 2017

   Abs HTML PDF

   With the growing popularity of Internet of Things (IoT) services being applied in several aspects of real-life applications, performance has become an important requirement. Meanwhile, the techniques for reliability enhancement such as virtual machine migration and recovery also have significant impact on end-to-end performance. This paper proposes a predictive approach of reliability-aware performance evaluation for recoverable IoT services using the modeling techniques of generalized stochastic Petri net (GSPN). Mathematical models formulating the dynamics of both server clusters and IoT systems are presented, and quantitative analyses of performance metrics are provided. Empirical experiments based on real-world data obtained from IoT services and cloud systems are conducted, and parameter settings as well as experimental results are discussed in detail.