A pronounced imbalance in GPU resources exists on campus, where some laboratories own underutilized servers while others lack the compute needed for AI research. GPU sharing can alleviate this disparity, while existing platforms typically rely on centralized oversight and persistent allocation models, conflicting with the voluntary and autonomous nature of academic resource ownership. We present GPUnion, a campus-scale GPU sharing platform enabling voluntary participation while preserving full provider autonomy. GPUnion incorporates three core mechanisms: i) container-based task dispatching and execution, ii) resource provider-first architecture, and iii) resilient execution featuring automatic checkpointing and migration. Case studies across multiple campus scenarios demonstrate 30% more GPU utilization improvement, 40% increase in interactive sessions, and 94% successful workload migration during provider departures.

The prominent paradigm of federated learning (FL) is increasingly being applied to emerging and cross-silo applications, particularly with edge computing systems serving as pivotal agents. However, this shift also renders FL training more energy and carbon intensive. To this end, we propose GreenFL, a carbon-aware FL training framework designed to systematically navigate the trade-offs between carbon emission, training accuracy, and training efficiency. GreenFL employs a hybrid training strategy that combines inter-group asynchronous training and intra-group synchronous training to mitigate the straggler effect caused by inefficient participants. In the overall design of the framework, we promote the participation of edge computing nodes with abundant renewable energy sources and implement strategic participant selection to balance carbon emissions and training accuracy. We prove the solvability of optimizing the selection strategy and provide an online greed-based solution based on penalty values and bipartite greedy algorithms. Through extensive data-driven experiments, we demonstrate that GreenFL can significantly improve the carbon efficiency of the entire FL procedure, while maintaining or exceeding state-of-the-art levels of training accuracy and efficiency.

Location-based mobile services, while improving user daily life, also raise significant privacy concerns in the sharing of location data. These trajectories indicate users’ traveling behavioural traces with rich semantics derived from open-source information. Behavioral-semantic analysis reveals users’ travelling motivations and underlying behavioral patterns. It contributes to attackers launching inferential attacks for behavior prediction, identity identification, or other privacy invasions, even when the location data is protected. It remains open to the issues of behavioral-semantic privacy-risk quantification and privacy-protection evaluation. This paper aims to reveal such semantic privacy risks of user behaviors arising from the publication of location trajectories in mobile scenarios. We formalize user semantic-mobility process to analyze his underlying behavior patterns. Then, we design semantic inference algorithms conditional on the released trajectory to reason about the observation-based likelihood of the user’s actual staying and transfer behaviours and behavioural-trace tracking. Extensive experiments with real-world data demonstrate their performance on inference accuracy and semantic similarity, offering a quantification criterion for deploying mobile privacy protection.

Many network attacks, like link flooding attacks (LFAs), heavily rely on network topology information. Therefore, network topology obfuscation has been applied to counteract network topology inference and prevent topology information leakage. One effective way is to scheme a fake topology intentionally for attackers to map out. Focusing on reducing the similarity between the real and fake topologies, however, existing methods cannot promise that critical information of the network, such as critical nodes and links, is well hidden. To this end, we propose a new topology obfuscation mechanism, namely ChameleonNet, to protect the critical topology information of a given network. Specifically, ChameleonNet achieves topology obfuscation through a two-stage operation: 1) generating fake topology and 2) deploying fake topology. Our experiments on three real-world and two large-scale generated network topologies demonstrate that ChameleonNet can effectively reduce similarity between inferred and real topologies by 31%-37% and reliably hide critical topology information in terms of multiple statistical metrics.

The geographically distributed edge servers can naturally draw power from nearby renewable energy (RE) generators. Complemented by the dynamic scheduling of energy storage batteries, edge service providers (ESPs) can thus build low- or even zero-carbon edge computing systems. Nevertheless, the distributed and heterogeneous nature of edge computing systems, as well as the limited information sharing among ESPs, leads to a more complex battery planning problem than that in cloud computing. The unpredictability of RE resources further complicates the problem, making conventional model-based approaches ineffective. To this end, we propose a multi-agent deep reinforcement learning (MADRL) approach for the independent decision making of individual ESPs. Particularly, MADRL takes privacy into account by ensuring that no sensitive information is disclosed among ESPs. For better model training, we further customize the invalid action masking and develop action transformation techniques based on segmented linear optimization. Extensive experiments demonstrate that, with our proposed approach, the overall carbon emission of edge computing systems can be significantly reduced (by over 60%) while maintaining acceptable operation costs in battery scheduling.

A content delivery network (CDN) aims to reduce the content delivery latency to end-users by using distributed cache servers. Nevertheless, deploying and maintaining cache servers on a large scale is very expensive. To solve this problem, CDN providers have developed a new content delivery strategy: allowing end-users’s IoT edge devices to share their storage/bandwidth resources. This new edge CDN platform must address two core questions: (1) how can we incentivize end users to share IoT devices? (2) how can we facilitate a safe and transparent content transaction environment for end users? This paper introduces SmartSharing, a new content delivery network solution to address these questions. In smartSharing, the over-the-top (OTT) IoT devices belonging to end-users are used as mini-cache servers. To motivate end users to share the idle devices and storage/bandwidth resources, SmartSharing designs the content delivery schedule and the pricing scheme based on game theory and machine learning algorithms (specifically, a tailored Expectation-Maximization (EM) algorithm). To facilitate content trading among end users, SmartSharing creates a secure and transparent transaction platform based on smart contracts in Ethereum. In addition, SmartSharing’s performance evaluation is through trace-driven simulations in the real world and a prototype using content metadata and the achieved pricing schemes. The evaluation results show that CDN providers, end users and content providers can all benefit from our SmartSharing framework.

Root cause analysis is the process of identifying the underlying factors responsible for triggering anomaly detection alarms. In the context of anomaly detection for Internet of Things (IoT) traffic, these alarms can be triggered by various factors, not all of which are malicious attacks. It is crucial to determine whether a malicious attack or benign operations cause an alarm. To address this challenge, we propose an innovative root cause analysis system called score-variational autoencoder (VAE), designed to complement existing IoT anomaly detection systems based on the federated learning (FL) framework. Score-VAE harnesses the full potential of the VAE network by integrating its training and testing schemes strategically. This integration enables Score-VAE to effectively utilize the generation and reconstruction capabilities of the VAE network. As a result, it exhibits excellent generalization, lifelong learning, collaboration, and privacy protection capabilities, all of which are essential for performing root cause analysis on IoT systems. We evaluate Score-VAE using real-world IoT trace data collected from various scenarios. The evaluation results demonstrate that Score-VAE accurately identifies the root causes behind alarms triggered by IoT anomaly detection systems. Furthermore, Score-VAE outperforms the baseline methods, providing superior performance in discovering root causes and delivering more accurate results.

Crowdsensing-based mobile Internet, while facilitating users’ daily life, also raises privacy concerns because of sharing user location trajectories. Combining with open-source network information, these trajectories reveal the semantics of users’ social behaviors in their travels, thus indicating their behavioral traces. Based on such social mobilities, attackers can explore users’ potential behavioral patterns and launch powerful behavioral-semantic inferential attacks for behavior prediction, identity identification, and threatening users’ location-related mobile privacy. Even through privacy protection, the released similar anonymous semantics may still bring significant privacy gains to such attacks. To the best of our knowledge, there is still no effective technique to counter such attacks and protect user behavioral semantics in mobile Internet services. To this end, this article proposes a posterior behavioral-semantic privacy-preserving solution, BSPri, by simulating the inferential attacks to eliminate the privacy risks associated with released traces. Specifically, we represent the logical association between semantic attributes and propose a similar semantic clustering and ranking method. Then, we formalize the user social-mobility stochastic process to characterize the privacy risks arising from the attacker’s observation of the released trajectory, and define a observation-based posterior privacy authentication criteria to filter anonymous semantics further. Finally, we generate synthetic trajectories with similar anonymity semantics, which bring attackers insignificant privacy gain, for users to participate in applications. Extensive experiments with the real-world data set demonstrate that our BSPri achieves an effective privacy-preserving performance, i.e., rigorous posterior-privacy constraint with limited data-availability loss, such as, distance 752 m$(47$-m closer, compared with our previous work MSP), direction deviation$39.5^{\circ }~(11.5^{\circ }$smaller), and semantic similarity$43.4\%~(8.4\%$closer).

While facilitating user daily lives, the booming development of mobile Internet services raises their privacy concerns because of the need to share travel trajectories. Due to the differences in access patterns and sensitive location attributes, behavioral semantics of user travel suffer from different degrees of leakage risks and have personalized privacy requirements. Semantic mobility-aware personalized privacy protection is still an open research issue in mobile scenarios. To this end, we propose a differentially semantic-generalized behavioral trajectory (DSG-BTra) for achieving privacy-preserving mobile Internet services. Specifically, we first explore the underlying behavioral patterns by formalizing user social mobility. Then, we evaluate the differential privacy sensitivity of user behavior to indicate the risks it faces. Finally, we generalize the behavioral semantics with a sensitivity-quantified strength and generate a DSG-BTra for the user to participate in mobile services. Extensive experiments with real-world data sets demonstrate DSG-BTra achieves flexible balance between privacy protection and application QoS, e.g., reducing the inference probability to 0.18–0.26 with a semantic similarity of 0.3–0.5.

The novel IoT-based data sensing and service mode promotes the booming development of crowdsensing-based mobile communication services (MCSs). MCS facilitates people’s daily lives by providing appropriate services according to the user’s mobile travels. These traveling trajectories, combined with open-source network information, reveal multimodal semantic information implicit in user mobility. Mining these mobile semantics contributes to understanding user mobility more sufficiently. It covers a wide spectrum of applications in mobile scenarios. For service providers, it improves the quality of their services. For mobile users, it helps to design a more rigorous privacy-preserving mechanism. For third-party platforms, such mobility analysis enhances their data management, analysis, and reusage. It has always been an open research issue in mobile computing. We are motivated to conduct a complete and comprehensive survey on semantic mining within the scope of MCS, forming a complete overview of mobile semantic perception. Specifically, we first review existing research works on feature selection. We classify them into five categories, depending on their representation forms. Then, we summarize the research on mobile semantic perception and cluster them to be three groups according to the digging depth of the represented semantics. To complete the overview, we also review the applications of learning algorithms and discuss the open opportunities and challenges for future works.

Geo-distributed edge data centers (EDCs) are expected to handle a large portion of tasks offloaded from cloud data centers for various emerging edge services. However, the high energy consumption and cost add a huge burden to edge service providers (ESPs). This presents a unique challenge as traditional energy-saving strategies applicable in cloud data centers fail to apply to EDCs, given the latency-sensitive nature of edge services. In response, we put forward an innovative electric vehicle (EV)-assisted edge computing architecture that leverages idle computing resources and stored energy of EVs. Our design aims to decrease energy expenditures for ESPs by choosing EVs with more economical service costs to handle a portion of the edge services during critical periods. We construct an energy cost-aware workload offloading model and discretize the original model into multiple small-scale solvable forms in both temporal and spatial dimensions. Furthermore, we reconfigure the Kuhn-Munkres algorithm to produce an online joint matching solution to counter QoS decline, generating a mutually advantageous situation for ESPs and EV participants. Upon experimentation with real-world traces, our design demonstrates a significant reduction in total energy cost (up to 31%) and offers considerable incentives for EV participants.

Each flow usually needs to traverse a specific service function chain (SFC), which is composed of multiple network functions implemented through virtualization technology or hardware, before reaching their destinations. All network functions are deployed across commodity nodes inside a network environment. Each flow needs to change its default routing path to visit the corresponding SFC correctly. These changed routing paths will cause network load imbalance. Therefore, an intelligent routing planning method is needed to balance the traffic load while satisfying various SFC requirements of different flows. In this paper, we propose to leverage SRv6, a new routing technology, to centrally plan the routing path for each flow with any SFC request. We then present a general model of the SFC planning problem (SFCP), planning flows’ routing paths to minimize the maximum link utilization of the network, and prove that the problem is NP-hard. For this reason, we transform the SFCP problem into a graph theory optimization problem and propose SFCPlanner, an online SFC planning method based on deep reinforcement learning. Moreover, we design the node mask and incremental training mechanisms to make SFCPlanner achieve better performance. The experiment results show that our SFCPlanner can solve the SFCP problem in large-scale networks more precisely. It can reduce the maximum link utilization by 32% compared with the benchmark algorithm while ensuring each flow traverses the correct SFC.

Clustered federated learning (CFL) is a promising solution to address the non-IID problem in the spatial domain for federated learning (FL). However, existing CFL solutions overlook the non-IID issue in the temporal domain and lack consideration of time efficiency. In this work, we propose a novel approach, calledClusterFLADS, which takes advantage of the false predictions of the inappropriate global models, together with knowledge of temperature scaling and catastrophic forgetting to reveal distributional similarities between the training data (of different clusters) and the test data. Additionally, we design an efficient feature extraction scheme by exploiting the role of each layer in a neural network's learning process. By strategically selecting model parameters and using PCA for dimensionality reduction,ClusterFLADSeffectively improves clustering speed. We evaluateClusterFLADSusing real-world IoT trace data in various scenarios. Our results show thatClusterFLADSaccurately and efficiently clusters clients, achieving a$100\%$true positive rate and low false positives across various data distributions in both the spatial and temporal domains.

The booming edge computing market that is supported by the edge cloud (EC) infrastructure has brought huge operating costs, mainly the energy cost, to edge service providers. The energy cost in form of electricity bills usually consists of energy charge and demand charge, and the demand charge based on peak power may account for a large proportion of the energy cost given a significant fluctuating power curve. In this work, we investigate the backup battery characteristics and electricity charge tariffs at ECs and explore the corresponding cost-saving potential. Specifically, we transform the backup battery group into distributed battery energy storage system (BESS) and strategically schedule the BESS to minimize the energy cost of service providers. We then propose a deep reinforcement learning (DRL) based approach to BESS charging/discharging in coping with the dynamic power demand and BESS state at each EC. To enable better decision-making and speed up agent training, we further design the customized invalid action masking (IAM) method and apply the prioritized experience replay (PER) scheme. The experiment results based on real-world EC power traces show that the proposed approach can reduce the demand charge and overall electricity bill by up to 27% and 13%, respectively.

Distributed learning has been widely adopted to train a global model from local data. However, its performance can be severely affected by stragglers. Recently, some research has been dedicated to resolving the straggler problem by adopting gradient coding, the essence of gradient coding is to solve the straggler problem by adding data redundancy. However, the large amount of data redundancy as well as computation and communication overhead that it brings is still hard to be resolved. Besides, the complexity of the encoding and decoding will increase linearly with the number of the local workers. To this end, in this paper, we design a lightweight coding method in the computing phase and seek to ensure fair transmission in the communication phase. Specifically, to tolerate stragglers in computing phase, we propose a two-stage dynamic coding scheme, part of the workers start computing the partial gradients from the data partitions assigned in the first stage, and the remaining workers for computation in the second stage is decided based on which workers have finished in the first stage. To further tolerate stragglers in the communication phase, a perturbed Lyapunov function is designed to maximize admission data balancing fairness as well as the throughput. The experimental result verifies the derived properties and demonstrates that our proposed solution can achieve a better performance for practical network parameters and benchmark data in terms of accuracy and resource utilization in the distributed learning system.

With the widespread use of Internet of Things (IoT) devices and the arrival of the 5G era, edge computing has become an attractive paradigm to serve end-users and provide better QoS. Many efforts have been paid to provision some merging public network services at the network edge. We reveal that it is very common that specific users call for private and isolated edge services to preserve data privacy and enable other security intentions. However, it still remains open to fulfill such kind of mixed requests in edge computing. In this article, we propose a cooperative edge computing framework, i.e., HyEdge, to offer both public and private edge services systematically. To fully exploit the benefits of this novel framework, we define the problem of optimal request scheduling over a given placement solution of hybrid edge servers to minimize the response delay. This problem is further modeled as a mixed integer non-linear programming problem (MINLP), which is typically NP-hard. Accordingly, we propose the partition-based optimization method, which can efficiently solve this NP-hard problem via the problem decomposition and the branch and bound strategies. We finally conduct extensive evaluations with a real-world dataset to measure the performance of our method. The results indicate that the proposed method achieves elegant performance with low computation complexity.

In Network Function Virtualization (NFV), a series of Virtual Network Functions (VNFs) organized in a specific order (called Service Function Chain, SFC) could offer an end-to-end network service for a network flow. Recently, with the new results of the exploration of VNF parallelism, hybrid SFC (SFC contains parallel VNFs) is proposed to reduce the SFC execution delay. However, it remains challenging and open to optimally embed multiple hybrid SFCs into the network. In this paper, we target at the optimal embedding problem of multiple hybrid SFCs with the purpose of minimizing the cost in an online scenario. Specifically, we propose SFT-Box, an online approach that can respond to hybrid SFC embedding requests in real-time. SFT-Box is designed to i) transform SFCs from the traditional sequential form to a standardized hierarchical Service Function Tree (SFT) form, ii) calculate and store the low-cost sub-solutions of embedding common SFTs, and iii) provide prompt solution response based on stored sub-solutions. To the best of our knowledge, this is the first work to address the online optimal embedding problem of multiple hybrid SFCs. With extensive evaluations, we demonstrate that, compared with the benchmark methods, SFT-Box can achieve up to 30% cost-saving and at least$22\times $latency reduction in enabling real-time response.

This paper explores resource allocation strategy in the Baidu Over The Edge system to enable mobile edge computing (MEC) datacenters to effectively support cloud-native services downstream to the network edge. There are many challenges to this issue. First, MEC datacenters are resource-constrained to fully meet resource demands. Second, previous works regard the resource requirements of each service as an indivisible unit, resulting in idle MEC resources, even if the resources can meet the demands of some microservices decoupled by the service. Third, they are confined to optimize the allocation for a single slot, failing to adapt to the dynamic demands. To improve resource utilization, we propose performability-aware resource allocation (PARA), a PARA on the edges for cloud-native services. It takes microservices as the unit of resource allocation and allows services to perform with degraded services when only part of microservices' demands are met. It also considers dependency among microservices, dynamic resource requirements, and resource supply characteristics of MEC and cloud. Performability is a unified performance-reliability measure for evaluating such degradable systems. To maximize the long-term overall performability, we model the resource optimization problem and then develop an online greedy heuristic algorithm. The algorithm predicts services' resource demands and then adapts the online allocation. The experimental results show that PARA reduces the reallocation overhead by 47.7%–53.6%, and improves the long-term overall performability by 23.14%–43.25% of existing state-of-the-art works.

Video streaming service is one of the most important IoT applications/services at the mobile end. To provide better services and earn more customers, the mobile video service providers have paid considerable attention to enhance end users’ Quality of Experience (QoE) in video streaming. As an important aspect of the mobile device, however, the battery power and its impacts on the mobile services were seldom concerned. According to our survey over 2000+ mobile users, the low battery power of mobile phones could cause the user to give up watching videos. To quantify the relationship between the battery power and user’s video abandoning probability (VAP), we first extract the VAP model from the collected survey data, leveraging a reversed accumulative histogram approach. Then, referring to the quantified VAP model, we presentEdgeSaver, an edge-assisted video transmission framework, which aims at maintaining a sustainable overall user retention rate for the service providers by reducing the power consumption of video playback at the mobile ends. Particularly, as the core component ofEdgeSaver, a low-power video scheduler is designed to strategically select user groups, such that the most profitable outcome can be achieved under the constraints of limited edge resources. With extensive experiments using a real-world data set, we demonstrate thatEdgeSavercan help the mobile video service provider improve the user retention rate by up to 30% and increase the average user viewing time by 20%.

The pervasive low-battery anxiety (LBA) among modern mobile users has a negative impact on users’ emotion and health, and such anxiety may directly lead to the loss of customers in power-hungry applications, e.g., video streaming. Despite its importance, LBA has not been thoroughly investigated due to the difficulty in quantitatively measuring LBA. To fill the gap, we present a quantitative model to measure the LBA of mobile users and design a tailored mechanism for LBA alleviation by saving display energy in video streaming. In specific, we first conduct a large-scale survey of 2000+ mobile users and strategically extract an empirical LBA model that captures the variation of users’ anxiety degree along with the battery power draining. Then, by exploiting the emerging edge computing paradigm, we propose a novel solution for low-power video streaming services (namely, LPVS) at the network edge. It aims to minimize the LBA of mobile users, by integrating the extracted LBA model with the energy-saving image/video content transforming techniques. To accommodate the heterogeneous LBA properties of mobile users under different circumstances, we develop an LBA model updating scheme by further considering the specific and local edge environments. The emulation results using real-world video watching traces demonstrate that LPVS can effectively alleviate mobile users’ LBA and prolong their video watching time (i.e., customer retention) by 39%.

The user demands to delay-sensitive applications have put forward new requirements for the mobile networks. Edge caching as a promising way is proposed to enhance the Quality of Service (QoS) for end users at the network edge. Given the widely distributed edge nodes, the content providers (CPs) usually prefer to integrate them with cooperative caching services, by forming a cache coalition. Although such a coalition could be beneficial as a whole, it neglects the profits of individual members, which is one major concern in forming the coalition itself. Besides, due to the poor scalability, the conventional cooperation scheme, which only considers fixed edge nodes, cannot adapt to the spatial and temporal imbalance of user requests. In this article, we tackle the problems of coalition establishment and profit allocation among the coalition members. Particularly, by adopting both fixed edge nodes and mobile vehicles as caching nodes, we propose a hybrid service provisioning framework and cooperative service caching and workload scheduling methods. To maximize the profits in managing the caching resources in the established coalition, we devise an optimization model with a mixed-integer programming (MIP), in which the QoS requirements of end users and caching capacities of each coalition member are also considered as constraints. In addition, based on the Hedonic game theory, we propose a dynamic coalition algorithm to guide each member to join or leave the coalition at each time slot out of its own profits. The experimental results demonstrate that compared to the cases only considering fixed caching nodes, our hybrid caching scheme can improve: 1) the overall profit of the coalition by 53% and 2) the average profit of individual participants by 42%, respectively.

As the development of new generation mobile communication technology, the mobile core network also needs to be upgraded by new network technologies, e.g., software defined networking (SDN) and network function virtualization (NFV). With NFV, virtual network functions (VNFs) can be deployed on commodity devices to support various network function requirements and attain system's flexibility and elasticity in network edge. Meanwhile, a set of selected VNFs are usually chained as a service function chain (SFC) to serve a given flow in a specified order. Since the devices have heterogeneous execution environments and the VNFs have various requirements, one fundamental challenge is how to embed SFC for each flow on the shared NFV infrastructure (NFVI) with the goal of minimizing the flow completion time. Furthermore, multiple flows always compete for resources of those devices hosting SFCs. In this general setting, there is an urgent need to study efficient scheduling mechanism to minimize the total completion time of all flows. In this paper, by jointly considering VNF placement and flow scheduling, we first formulate this problem as an integer programming problem, and further prove that it is NP-hard in general case. We then design a PDG method to find the optimal solution in single flow case and an LRD method to achieve a high-quality feasible solution in multiple flows case. The extensive experiment results indicate that our LRD method can reduce the total completion time of all flows by 22.04, 60.99 and 39.95, percent against three compared methods, respectively.

Many organizations nowadays have multiple sites at different geographic locations. Typically, transmitting massive data among these sites relies on the interconnection service offered by ISPs. Segment Routing over IPv6 (SRv6) is a new simple and flexible source routing solution which could be leveraged to enhance interconnection services. Compared to traditional technologies, e.g., physical leased lines and MPLS-VPN, SRv6 can easily enable quick-launched interconnection services and significantly benefit from traffic engineering with SRv6-TE. To parse the SRv6 packet headers, however, hardware support and upgrade are needed for the conventional routers of ISP. In this paper, we study the problem of SRv6 incremental deployment to provide a more balanced interconnection service from a traffic engineering view. We formally formulate the problem as an SRID problem with integer programming. After transforming the SRID problem into a graph model, we propose two greedy methods considering short-term and long-term impacts with reinforcement learning, namely GSI and GLI. The experiment results using a public dataset demonstrate that both GSI and GLI can significantly reduce the maximum link utilization, where GLI achieves a saving of 59.1% against the default method.

OpenFlow switches are being deployed in SDN to enable a wide spectrum of non-traditional applications. As a promising alternative to brutal force TCAMs, FPGA-based packet classification is being actively investigated. However, none of the existing FPGA designs can achieve high performance on both search and update for large-scale rule sets. To address this issue, we propose TcbTree, an FPGA-based algorithmic scheme for packet classification. Specifically, at the algorithmic side, i) a two-stage framework consisting of heterogeneous algorithms is proposed, where most rules can be mapped into several balanced trees without rule replications, ii) for the remaining few rules, a centralized TSS (Tuple Space Search) architecture together with a real-time feedback scheme is designed to enhance the efficiency of TSS search on FPGA, and iii) a tree dilution method is designed to equalize rule distribution in trees, so that the latency of tree search can be reduced. At the hardware side, i) an efficient data structure set is designed to convert tree traversal to addressing process, which breaks the constraints of limited tree depth and imbalanced node distribution, and ii) distinct from fully pipelined designs, multiple levels of parallelism are efficiently explored with multi-core, multi-search-engine and coarse-grained pipelines herein. Experimental results using ClassBench show that, with the implementation of TcbTree on FPGA, the average classification throughputs for 1k, 10k, 32k and 100k rule sets achieve 788.8 MPPS, 404.3 MPPS, 237 MPPS and 41.8 MPPS, respectively, and the update throughput for all benchmark rule sets is above 1 MUPS.

Many hardware-based security middleboxes have been deployed in the networks to defend against different threats. However, these hardware middleboxes are hard to upgrade or migrate. The emergence of network functions virtualization (NFV), which realizes various security functions in the form of virtual network functions (VNFs), brings many benefits to network security. To improve the security level further, several VNFs are coordinated in a pre-defined order to form service function chains (SFCs). It is expected that the SFCs are embedded properly with low cost, including the VNF setup cost and the flow routing cost. In this paper, we find that when an SFC is required by multiple flows for the identical network security threats, the total cost could be reduced by embedding an aggregated service function tree (ASFT) instead of multiple independent SFCs. We formally characterize the integer programming model of this problem and prove that it is NP-hard. Then we propose a performance-guaranteed approximation algorithm and prove that the algorithm could find the optimal solution in a special case. Extensive experiments indicate that our method can reduce the total cost by$22.0\%$and$24.1\%$against two compared algorithms, respectively.

Live streaming services have gained extreme popularity in recent years. Due to the spiky traffic patterns of live videos, utilizing distributed edge servers to improve viewers' quality of experience (QoE) has become a common practice nowadays. Nevertheless, the current client-driven content caching mechanism does not support pre-caching from the cloud to the edge, resulting in a considerable amount of cache misses in live video delivery. By jointly considering the features of live videos and edge servers, we propose PLVER, a proactive live video push scheme to address the cache miss problem in live video delivery. Specifically, PLVER first conducts a one-to-multiple stable allocation between edge clusters and user groups to balance the load of live traffic over the edge servers. It then adopts proactive video replication algorithms to speed up video replication among the edge servers. We conduct extensive trace-driven evaluation, covering 0.3 million Twitch viewers and more than 300 Twitch channels. The results demonstrate that with PLVER, edge servers can carry 28 and 82 percent more traffic than the auction-based replication (ABR) method and the caching on requested time (CORT) method, respectively.

Existing Internet routing protocols much focus on providing interconnection service for independent autonomous systems (ASes) rather than end-to-end low latency transmission. Nowadays, a growing number of applications and platforms have high requirements for low latency. However, developing new routing protocols in the wide area network that provides low latency routing service is very challenging, and remains an open problem due to the obstacles of compatibility, feasibility, scalability and efficiency. On the other hand, the ignorance of latency performance results in triangle inequality violations (TIV). In this paper, we leverage TIV and a new routing technology, SRv6, to build a new distributed routing protocol, SRv6 underlay federation (SRUF), which aims to provide low-latency routing services in network core. We design a novel method to find alternative paths with lower latency between any pair of ASes in SRUF. This method can achieve high scalability as it incurs only$O(n)$bandwidth overhead in each member of SRUF. SRv6 is then employed to steer the flows along the selected indirect low-latency paths, while keeping compatibility to legacy routing systems. The experimental results with realworld datasets demonstrate that SRUF can effectively reduce the average end-to-end delay by 5.4% ~ 58.9%.

The mobile network operators are upgrading their network facilities and shifting to the 5G era at an unprecedented pace. The huge operating expense (OPEX), mainly the energy consumption cost, has become the major concern of the operators. In this work, we investigate the energy cost-saving potential by transforming the backup batteries of base stations (BSs) to a distributed battery energy storage system (BESS). Specifically, to minimize the total energy cost, we model the distributed BESS discharge/charge scheduling as an optimization problem by incorporating comprehensive practical considerations. Then, considering the dynamic BS power demands in practice, we propose a deep reinforcement learning (DRL) based approach to make BESS scheduling decisions in real-time. The experiments using real-world BS deployment and traffic load data demonstrate that with our DRL-based BESS scheduling, the peak power demand charge of BSs can be reduced by up to 26.59%, and the yearly OPEX saving for 2,282 5G BSs could reach up to US$185,000.

Blockchain-powered edge computing (BEC) is a promising extension to strengthen the security and the trustworthiness among collaborative edge clouds for delivering computation-intensive and delay-sensitive services in the environments of IoT and 5G. A fundamental challenge is how to respond to the maximum number of IoT requests at the network edge instead of the remote cloud. Although some work has been done to consider service provisioning and request scheduling in collaborative edge clouds, they assume that a single service is used to respond to each request. This assumption, however, is not practical to meet the demand of emerging IoT applications. In reality, the request needs to call a set of services with a chain-based structure. To tackle this challenge, in this article, we first propose a chain-based service request model for emerging IoT applications and further study the joint service provisioning and request scheduling problem for chain-based service requests at the network edge. We characterize this problem as an integer linear programming (ILP) model and prove the NP-hardness of this joint optimization problem. Furthermore, we prove that the related problem is of approximate submodularity with an approximation ratio guarantee. Finally, a novel two-stage optimization (TSO) scheme is proposed, and the results of extensive experiments show the efficiency and the effectiveness of the TSO scheme.

Synthesizing a fake trajectory with consistent lifestyle and meaningful mobility as the actual one is the most popular way to protect the location privacy in trajectory sharing. Recent location privacy preservation shows a strong personalized requirement from the mobile semantics between users and locations. However, the existing techniques cannot fully satisfy such personalized requirements, resulting in either overprotection or underprotection. It remains open to characterize and quantify the personalized requirement for location privacy preservation. In this article, we propose a mobile semantic-aware privacy model, named MSP. Specifically, we first characterize a new kind of user-related mobile semantic on-location set by constructing a hierarchical semantic tree, according to the user’s roles at locations. Then, a dedicated approach is proposed to evaluate the location’s privacy sensitivity and integrate it into the user-related mobile semantic. Finally, an adaptive privacy-preserving mechanism, MSP, is developed, fully considering the personalized requirement from both the user and the location. With this model in place, mobile semantic-aware synthetic trajectories are constructed adaptively. Extensive experiments with a real-world data set demonstrate that our MSP model can achieve an effective and flexible balance between the personalized privacy preservation and the data availability of synthetic trajectories.

The emergence of edge computing can effectively tackle the problem of large transmission delays caused by the long-distance between user devices and remote cloud servers. Users can offload tasks to the nearby edge servers to perform computations, so as to minimize the average task response time through effective task dispatching and scheduling methods. However: 1) in the task dispatching phase, the dynamic features of network conditions and server loads make it difficult for the offloaded tasks to select the optimal edge server and 2) in the task scheduling phase, each edge server may face a large number of offloading tasks to schedule, resulting in long average task response time, or even severe task starvation. In this article, we propose an online task dispatching and fair scheduling method OTDS to tackle the above two challenges, which combines online learning (OL) and deep reinforcement learning (DRL) techniques. Specifically, using an OL approach, OTDS performs real-time estimating of network conditions and server loads, and then dynamically assigns tasks to the optimal edge servers accordingly. Meanwhile, at each edge server, by combing the round-robin mechanism with DRL, OTDS is able to allocate appropriate resources to each task according to its time sensitivity and achieve high efficiency and fairness in task scheduling. Evaluation results show that our online method can dynamically allocate network resources and computing resources to those offloaded tasks according to their time-sensitive requirements. Thus, OTDS outperforms the existing methods in terms of the efficiency and fairness on task dispatching and scheduling by significantly reducing the average task response time.

City-wide package delivery becomes popular due to the dramatic rise of online shopping. In order to speed up the package delivery process without increasing the delivery cost, a promising system has been proposed, which leverages the crowdsourced taxis. Many efforts have been done on this novel system in recent literature. However, a fundamental problem still remains open, i.e., measuring the maximum capacity of taxi-based logistics at the urban scale. In this paper, we first propose an accurate and efficient measurement mechanism to tackle this problem in the Non-stop package delivery method. The basic idea is to construct a spatial-temporal graph according to the passenger demands and calculate the maximum urban capacity by combining the results of several carefully designed max-flow problems. Then, we expand our measurement mechanism to be used in other taxi-based package delivery methods after a few adaptations, including the One-hop method and the Stop-and-wait method. At last, we evaluate our measurement mechanism and compare the maximum urban capacity of various package delivery methods with a real-world dataset from an online taxi-taking platform.

The pervasive low-battery anxiety (LBA) among modern mobile users has caused negative impacts on users' emotion and health, and such anxiety may directly lead to loss of customers in power-hungry applications, e.g., video streaming. Despite its importance, LBA has not been thoroughly investigated due to the difficulty in quantitatively measuring LBA. To fill the gap, we present a quantitative model to measure the LBA among mobile users and design a tailored mechanism to alleviate it via display energy saving in video streaming. In specific, we first conduct a large-scale user survey among 2000+ mobile users and strategically extract an empirical LBA model that captures the variation of user's anxiety degree along with the battery power draining. Then, by exploiting the emerging edge computing paradigm, we propose LPVS, a novel solution for low-power video streaming service at the network edge. It aims to minimize the LBA of mobile users, by integrating the extracted LBA model with the energy-saving image/video content transforming techniques. The emulation results using real-world video watching traces demonstrate that, LPVS can effectively alleviate mobile users' LBA and prolong the low-battery users' video watching time (i.e., customer retention) by 39%.

Recent years have seen a rapidly increasing traffic demand for HTTP-based high-quality live video streaming. The surging traffic demand, as well as the real-time property of live videos, make it challenging for content delivery networks (CDNs) to guarantee the Quality-of-Experiences (QoE) of viewers. The initial video segment (IVS) of live streaming plays an important role in the QoE of live viewers, particularly when users require fast join time and smooth view experience. State-of-the-art research on this regard estimates network throughput for each viewer and thus may incur a large overhead that offsets the benefit. To tackle the problem, we propose Rldish, a scheme deployed at the edge CDN server, to dynamically select a suitable IVS for new live viewers based on Reinforcement Learning (RL). Rldish is transparent to both the client and the streaming server. It collects the real-time QoE observations from the edge without any client-side assistance, then uses these QoE observations as real-time rewards in RL. We deploy Rldish as a virtualized network function (VNF) in a real HTTP cache server, and evaluate its performance using streaming servers distributed over the world. Our experiments show that Rldish improves the state- of-the-art IVS selection scheme w.r.t. the average QoE of live viewers by up to 22%.

City-wide package delivery has become popular due to the dramatic rise of online shopping. It places a tremendous burden on the traditional logistics industry, which relies on dedicated couriers and is labor-intensive. Leveraging the ridesharing systems is a promising alternative, yet existing solutions are limited to one-hop ridesharing or need consignment warehouses as relays. In this paper, we propose a new package delivery scheme which takes advantage of multi-hop ridesharing and is entirely consignment free. Specifically, a package is assigned to a taxi which is guided to deliver the package all along to its destination while transporting successive passengers. We tackle it with a two-phase solution, named PPtaxi. In the first phase, we use the Multivariate Gaussian distribution and Bayesian inference to predict the passenger orders. In the second phase, both the computation efficiency and solution effectiveness are considered to plan package delivery routes. We evaluate PPtaxi with a real-world dataset from an online taxi-taking platform and compare it with multiple benchmarks. The results show that the successful delivery rate of packages with our solution can reach 95 percent on average during the daytime, and is at most 46.9 percent higher than those of the benchmarks.

Edge computing is a promising computing paradigm by pushing the cloud service to the network edge. To this end, edge infrastructure providers (EIPs) need to bring computation and storage resources to the network edge and allow edge service providers (ESPs) to provision latency-critical services for end users. Currently, EIPs prefer to establish a series of private edge-computing environments to serve specific requirements of users. This kind of resource provisioning mechanism severely limits the development and spread of edge computing in serving diverse user requirements. In this paper, we propose an integrated resource provisioning model, namededge federation, to seamlessly realize the resource cooperation and service provisioning across standalone edge computing providers and clouds. To efficiently schedule and utilize the resources across multiple EIPs, we systematically characterize the provisioning process as a large-scale linear programming (LP) problem and transform it into an easily solved form. Accordingly, we design a dynamic algorithm to tackle the varying service demands from users. We conduct extensive experiments over the base station networks in Toronto. Compared with the fixed contract model and multihoming model, edge federation can reduce the overall costs of EIPs by 23.3% to 24.5%, and 15.5% to 16.3%, respectively.

Usually, a data flow needs to traverse a series of network functions, which is called a service function chain (SFC), before reaching its destination. The emergence of network function virtualization (NFV) makes the embedding solution of the SFC flexible as far as the deployment location is concerned. When providers embed the SFC into a substrate network, they will hope to minimize the setup cost of the SFC and link connection cost toward clients. For unicast, since there is one path connecting the source node to the destination node, all functions of the SFC are just needed to be deployed along the path. However, when embedding the SFC for a multicast task, the topology of the SFC may change because the function deployment locations have impacts on the traffic delivery cost. Thus, a service function tree (SFT) may be a better choice. Given the huge space of SFT embedding solutions, however, it is extremely hard to find the optimal one such that the total traffic delivery cost is minimized. In this paper, we tackle the optimal SFT embedding problem in the NFV enabled multicast task. Specifically, we formally define the problem and formulate it with an integer linear programming (ILP), which turns out to be NP-hard. Then, a two-stage method is proposed to deal with the problem with an approximation ratio of$1+\rho $, where$\rho $is the best approximation ratio of Steiner tree and can be as small as 1.39. With extensive experimental evaluations, we demonstrate that by applying our SFT embedding solution, the delivery cost of multicast traffic can be reduced by 22.05% at most against three benchmarks.

Driven by the visions of Internet of Things and 5G communications, the edge computing systems integrate computing, storage, and network resources at the edge of the network to provide computing infrastructure, enabling developers to quickly develop and deploy edge applications. At present, the edge computing systems have received widespread attention in both industry and academia. To explore new research opportunities and assist users in selecting suitable edge computing systems for specific applications, this survey paper provides a comprehensive overview of the existing edge computing systems and introduces representative projects. A comparison of open-source tools is presented according to their applicability. Finally, we highlight energy efficiency and deep learning optimization of edge computing systems. Open issues for analyzing and designing an edge computing system are also studied in this paper.

The content delivery network (CDN) intensively uses cache to push the content close to end users. Over both traditional Internet architecture and emerging cloud-based framework, cache allocation has been the core problem that any CDN operator needs to address. As the first step for cache deployment, CDN operators need to discover or estimate the distribution of user requests in different geographic areas. This step results in a statistical spatial model for the user requests, which is used as the key input to solve the optimal cache deployment problem. More often than not, the temporal information in user requests is omitted to simplify the CDN design. In this paper, we disclose that the spatial request model alone may not lead to truly optimal cache deployment and revisit the problem by taking the dynamic traffic demands into consideration. Specifically, we model the time-varying traffic demands and formulate the distributed cache deployment optimization problem with an integer linear program (ILP). To solve the problem efficiently, we transform the ILP problem into a scalable form and propose a greedy diagram to tackle it. Via experiments over the North American ISPs points of presence (PoPs) network, our new solution outperforms traditional CDN design method and saves the overall delivery cost by 16% to 20%. We also study the impact of various traffic demand patterns to the CDN design cost, via experiments with both real-world traffic demand patterns and extensive synthetic trace data.

In network traffic engineering, multicast is designed to deliver the same content from a single source to a group of destinations. Recently, NFV enabled multicast has been developed by deploying virtual network functions (VNFs) over the target network. To fulfill the multicast task with a service function chain (SFC) requirement, a service function tree (SFT) embedded in the shared multicast tree has to be built. Given the huge space of SFT embedding solutions, however, it is extremely hard to find the optimal one such that the total traffic delivery cost is minimized. In this paper, we tackle the optimal SFT embedding problem in NFV enabled multicast task. Specifically, we formally define the problem and formulate it with an integer linear programming (ILP), which turns out to be NP-hard. Then, a two-stage algorithm is proposed to deal with the problem with an approximation ratio of 1+p, where p is the best approximation ratio of Steiner tree and can be as small as 1.39. With extensive experimental evaluations, we demonstrate that by applying our SFT embedding solution, the cost saving of multicast traffic delivery can be up to 22.41%, compared with the random SFT embedding strategy.

Nowadays, more and more content providers (CPs) use multiple content delivery networks (CDNs) to deliver their content (a.k.a. content multihoming). Since the decisions on which CDN to use are made by the CP or by a CDN broker based on their local view of network conditions, content multihoming still has much room to improve for a better content delivery performance. In addition, content multihoming may negatively impact CDN vendors since in the price competition they are enforced to lower content delivery price to attract CPs to use their CDNs. To build a better CDN ecosystem, multi-CDN federation has been proposed to interconnect standalone CDNs. The real-world implementation of CDN interconnection (CDNI), however, poses significant technical obstacles not easy to solve in the short term. In order to improve the content delivery performance under current multi-CDN strategies, in this paper, we propose a feasible and efficient solution to multi-CDN, termed as CDN semi-federation, which can better schedule and utilize the resources from multiple CDNs without requiring full CDNI. The benefit of our solution comes from an effective optimization algorithm which reshapes the patterns of traffic from multiple CPs delivered over multipe CDN Points of Presence (PoPs). Experiments across North American and European ISP PoP networks demonstrate that, compared with current multi-CDN solutions, CDN semi-federation can reduce the content delivery latency by around 20% during peak traffic hours.

Network Function Virtualization (NFV) is an emerging technology, which enables service agility, flexibility and cost reduction by replacing traditional hardware middleboxes with Virtual Network Functions (VNFs) running on general-purpose servers. Service Function Chain (SFC) constitutes an end-to-end service by organizing a series of VNFs in a specific order. Particularly, hybrid SFC (SFC with parallel VNFs) is proposed to much reduce the traffic delay in sequential SFCs. Nevertheless, how to strategically select VNF instances and links in hybrid SFC embedding remains an open problem. In this paper, we target at the cost minimization and address the optimal hybrid SFC embedding problem. Specifically, we first develop a novel abstraction model for the hybrid SFC with Directed Acyclic Graph (DAG), which helps convert diverse hybrid SFCs to the standardized DAG-SFC form. Then, we formulate the optimal DAG-SFC embedding problem as an integer optimization model and propose a greedy method (called BBE) to solve the NP-hard problem. MBBE method is developed upon BBE method to further cut down the computation complexity in model solving. Extensive simulation results demonstrate the effectiveness of our approach for cost reduction in hybrid SFC embedding.

In shared bike-and-ride transit systems, commuters use shared bicycles for last-mile transport between transit stations and home, and between transit stations and work locations. This requires pools of bicycles to be located near each transit stop where commuters can drop off and pick up shared bikes. We study the optimal sizing of such bicycle pools. While various problems related to vehicle pool sizing have been studied before, to the best of our knowledge this is the first paper that considers a multimodal transportation system with a regularly scheduled public transportation backbone and shared bicycles for the first and last mile. We present two solutions that guarantee bicycle availability with high probability, and we empirically verify their effectiveness using Monte Carlo simulations. Compared to a baseline solution, our techniques reduce the size of the bikeshare pool at the public transit station from 39% to 75% in the tested scenarios.

The ever-increasing power consumption of datacenters has eaten up a large portion of their profit. One possible solution is to charge datacenter users for their actual power usage. However, it poses a great technical challenge as the power of VMs co-existing in a physical machine cannot be measured directly. It is thus critical to develop a fair method to disaggregate the power of a physical machine to individual VMs. We tackle the above challenge by modeling the power disaggregation problem as a cooperative game and propose non-deterministic Shapley value to discover the fair power share of VMs (in the sense of satisfying four desired axiomatic principles), while compensating the negative impact of VM power variation. We demonstrate that the results from existing power model-based solution can deviate from the "ground truth" by 25.22% ~46.15%. And compared with the exact Shapley value, our non-deterministic Shapley value can achieve less than 5% error for 90% of the time.

The content delivery network (CDN) intensively uses cache to push the content close to end users. Over both traditional Internet architecture and emerging cloud-based framework, cache allocation has been the core problem that any CDN operator needs to address. As the first step for cache deployment, CDN operators need to discover or estimate the distribution of user requests in different geographic areas. This step results in a statistical spatial model for the user requests, which is used as the key input to solve the optimal cache deployment problem. More often than not, the temporal information in user requests is omitted to simplify the CDN design. In this paper, we disclose that the spatial request model alone may not lead to truly optimal cache deployment. By considering the temporal information in user requests, we provide a dynamic traffic based solution to this broadly studied problem. Via experiments over the North American ISPs Points of Presence (PoPs) network, our new solution outperforms traditional CDN design method and saves the overall delivery cost by 16% to 20%.

Fine-grained power monitoring, which refers to power monitoring at the server level, is critical to the efficient operation and energy saving of datacenters. Fined-grained power monitoring, however, is extremely challenging in legacy datacenters that host server systems not equipped with power monitoring sensors. Installing power monitoring hardware at the server level not only incurs high costs but also complicates the maintenance of high-density server clusters and enclosures. In this paper, we present a zero-cost, purely software-based solution to this challenging problem. We use a novel technique of non-intrusive power disaggregation (NIPD) that establishes power mapping functions (PMFs) between the states of servers and their power consumption, and infer the power consumption of each server with the aggregated power of the entire datacenter. The PMFs that we have developed can support both linear and nonlinear power models via the state feature transformation. To reduce the training overhead, we further develop adaptive PMFs update strategies and ensure that the training data and state features are appropriately selected. We implement and evaluate NIPD over a real-world datacenter with 326 nodes. The results show that our solution can provide high precision power estimation at both rack level and server level. In specific, with PMFs including only two nonlinear terms, our power estimation i) at rack level has mean relative error of 2.18 percent, and ii) at server level has mean relative errors of 9.61 and 7.53 percent corresponding to the idle and peak power, respectively.

Role mining has been extensively used to automatically generate roles for role-based access control. Nevertheless, the two core problems in role mining, role minimization and edge concentration, are both NP-hard. While many approximate algorithms have been developed to solve the problems, experimental tests disclose that no algorithm clearly outperforms the others in both role minimization and edge concentration. The performance results highly depend on the data set under study. To determine the right role mining algorithm, a trial-and-error approach is time consuming due to the computational overhead in mining large data set. We tackle the problem from a fresh angle. Instead of developing fast role mining algorithms, we adopt a data-centric approach that quickly estimates the bounds on optimal role mining results without actually running any role mining algorithm. Based on the inherent features of the data set, the approach can also determine whether it is easy to achieve both role minimization and edge concentration, and if not, which direction, role minimization or edge concentration, that role mining could move toward further.

Non-intrusive appliance load monitoring (NIALM) helps identify major energy guzzlers in a building without introducing extra metering cost. It motivates users to take proper actions for energy saving and greatly facilitates demand response (DR) programs. Nevertheless, NIALM of large-scale appliances is still an open challenge. To pursue a scalable solution to energy monitoring for contemporary large-scale appliance groups, we propose a distributed metering platform and use parallel optimization for semi-intrusive appliance load monitoring (SIALM). Based on a simple power model, a sparse switching event recovering (SSER) model is established to recover appliance states from their aggregated load data. Furthermore, the sufficient conditions for unambiguous state recovery of multiple appliances are presented. By considering these conditions as well as the electrical network topology constraint, a minimum number of meters are obtained to correctly recover the energy consumption of individual appliances. We evaluate the performance of both SIALM and NIALM with real-world trace data and synthetic data. The results demonstrate that with the help of a small number of meters, the SIALM approach significantly improves the accuracy of energy disaggregation for large-scale appliances.

Fine-grained power monitoring, which refers to power monitoring at the server level, is critical to the efficient operation and energy saving of datacenters. Fined-grained power monitoring, however, is extremely challenging in legacy datacenters that host server systems not equipped with power monitoring sensors. Installing power monitoring hardware at the server level not only incurs high costs but also complicates the maintenance of high-density server clusters and enclosures. In this paper, we present a zero-cost, purely software-based solution to this challenging problem. We use a novel technique of non-intrusive power disaggregation (NIPD) that establishes power mapping functions (PMFs) between the states of servers and their power consumption, and infer the power consumption of each server with the aggregated power of the entire datacenter. We implement and evaluate NIPD over a real-world datacenter with 326 nodes. The results show that our solution can provide high precision power estimation at the rack level, with mean relative error of 2.63%, and the server level, with mean relative error of 10.27% and 8.17% for the estimation of idle power and peak power, respectively.

Load curve data in power systems refers to users' electrical energy consumption data periodically collected with meters. It has become one of the most important assets for modern power systems. Many operational decisions are made based on the information discovered in the data. Load curve data, however, usually suffers from corruptions caused by various factors, such as data transmission errors or malfunctioning meters. To solve the problem, tremendous research efforts have been made on load curve data cleansing. Most existing approaches apply outlier detection methods from the supply side (i.e., electricity service providers), which may only have aggregated load data. In this paper, we propose to seek aid from the demand side (i.e., electricity service users). With the help of readily available knowledge on consumers' appliances, we present an appliance-driven approach to load curve data cleansing. This approach utilizes data generation rules and a Sequential Local Optimization Algorithm (SLOA) to solve the Corrupted Data Identification Problem (CDIP). We evaluate the performance of SLOA with real-world trace data and synthetic data. The results indicate that, comparing to existing load data cleansing methods, such as B-spline smoothing, our approach has an overall better performance and can effectively identify consecutive corrupted data. Experimental results also show that our method is robust in various tests.