Texture recognition is a challenging visual task since its multiple primitives or attributes can be perceived from the texture image under different spatial contexts. Existing approaches predominantly built upon CNN incorporate rich local descriptors with orderless aggregation to capture invariance to the spatial layout. However, these methods ignore the inherent structure relation organized by primitives and the semantic concept described by attributes, which are critical cues for texture representation. In this paper, we propose a novel Multiple Primitives and Attributes Perception network (MPAP) that extracts features by modeling the relation of bottom-up structure and top-down attribute in a multi-branch unified framework. A bottom-up process is first proposed to capture the inherent relation of various primitive structures by leveraging structure dependency and spatial order information. Then, a top-down process is introduced to model the latent relation of multiple attributes by transferring attribute-related features between adjacent branches. Moreover, an augmentation module is devised to bridge the gap between high-level attributes and low-level structure features. MPAP can learn representation through jointing bottom-up and top-down processes in a mutually reinforced manner. Experimental results on six challenging texture datasets demonstrate the superiority of MPAP over state-of-the-art methods in terms of accuracy, robustness, and efficiency.

The rapidly expanding neural network models are becoming increasingly challenging to run on a single device. Hence, model parallelism over multiple devices is critical to guaranteeing the efficiency of training large models. Recent proposals either have long processing time or poor performance. Therefore, we propose Mercury, a fast framework for optimizing device placement for large models. Mercury employs a simple but efficient model parallelization strategy in the baseline measurement, and generates placement policies through a series of scheduling algorithms. We conduct experiments to deploy and evaluate Mercury on numerous large models. The results show that Mercury not only reduces the placement policy generation time by 26.4% but also improves the model throughput by 218.5% compared to the most advanced methods.

Camouflaged object detection is a challenging visual task since the appearance and morphology of foreground objects and background regions are highly similar in nature. Recent CNN-based studies gradually integrated the high-level semantic information and the low-level local features of images through hierarchical and progressive structures to achieve camouflaged object detection. However, these methods ignore thespatial statistical propertiesof the local context, which is a critical cue for distinguishing and describing camouflaged objects. To address this problem, we propose a novel Deep Texton-Coherence Network (DTC-Net) that leverages the spatial organization of textons in the foreground and background regions as discriminative cues for camouflaged object detection. Specifically, a Local Bilinear module (LB) is devised to obtain the robust representation of texton to trivial details and illumination changes, by replacing the classic first-order linearization operations with bilinear second-order statistical operations in the convolution process. Next, these texton representations are associated with a Spatial Coherence Organization module (SCO) to capture irregular spatial coherence via a deformable convolutional strategy, and then the descriptions of the textons extracted by the LB module are used as weights to suppress features that are spatially adjacent but have different representations. Finally, the texton-coherence representation is integrated with the original features at different levels to achieve camouflaged object detection. Evaluation on the three most challenging camouflaged object detection datasets demonstrats the superiority of the proposed model when compared to the state-of-the-art methods. Furthermore, our ablation studies and performance analyses demonstrate the effectiveness of the texton-coherence module.

Reasoning on knowledge graph (KG) has been studied for explainable recommendation due to its ability of providing explicit explanations. However, current KG-based explainable recommendation methods unfortunately ignore the temporal information (such as purchase time, recommend time, etc.), which may result in unsuitable explanations. In this work, we propose a novel Time-aware Path reasoning for Recommendation (TPRec for short) method, which leverages the potential of temporal information to offer better recommendation with plausible explanations. First, we present an efficient time-aware interaction relation extraction component to construct collaborative knowledge graph with time-aware interactions (TCKG for short), and then we introduce a novel time-aware path reasoning method for recommendation. We conduct extensive experiments on three real-world datasets. The results demonstrate that the proposed TPRec could successfully employ TCKG to achieve substantial gains and improve the quality of explainable recommendation.

Social bots are referred to as the automated accounts on social networks that make attempts to behave like humans. While Graph Neural Networks (GNNs) have been massively applied to the field of social bot detection, a huge amount of domain expertise and prior knowledge is heavily engaged in the state-of-the-art approaches to design a dedicated neural network architecture for a specific classification task. Involving oversized nodes and network layers in the model design, however, usually causes the over-smoothing problem and the lack of embedding discrimination. In this article, we propose RoSGAS , a novel R einf o rced and S elf-supervised G NN A rchitecture S earch framework to adaptively pinpoint the most suitable multi-hop neighborhood and the number of layers in the GNN architecture. More specifically, we consider the social bot detection problem as a user-centric subgraph embedding and classification task. We exploit the heterogeneous information network to present the user connectivity by leveraging account metadata, relationships, behavioral features, and content features. RoSGAS uses a multi-agent deep reinforcement learning (RL), 31 pages. mechanism for navigating the search of optimal neighborhood and network layers to learn individually the subgraph embedding for each target user. A nearest neighbor mechanism is developed for accelerating the RL training process, and RoSGAS can learn more discriminative subgraph embedding with the aid of self-supervised learning. Experiments on five Twitter datasets show that RoSGAS outperforms the state-of-the-art approaches in terms of accuracy, training efficiency, and stability and has better generalization when handling unseen samples.

Relational triple extraction is challenging for its difficulty in capturing rich correlations between entities and relations.Existing works suffer from 1) heterogeneous representations of entities and relations, and 2) heterogeneous modeling of entity-entity interactions and entity-relation interactions.Therefore, the rich correlations are not fully exploited by existing works.In this paper, we propose UniRel to address these challenges.Specifically, we unify the representations of entities and relations by jointly encoding them within a concatenated natural language sequence, and unify the modeling of interactions with a proposed Interaction Map, which is built upon the off-the-shelf self-attention mechanism within any Transformer block.With comprehensive experiments on two popular relational triple extraction datasets, we demonstrate that UniRel is more effective and computationally efficient.The source code is available at https://github.com/wtangdev/UniRel.

Person identification based on Electroencephalogram (EEG) signals has been studied through feature engineering, channel selection methods, and non-neural-network/deep models. Existing studies map signals from 3-D distributed channels into channel-time matrices, leading to spatial information loss. In this paper, we design an EEG-based person identification model EPI-CGAN, which combines the Convolution Neural Network (CNN) with conditional generative adversarial network (CGAN). More specifically, we map handcrafted temporal features into a 3-D feature tensor and apply CNN to extract spatial features. We leverage both handcrafted features and extracted features to improve model performance and robustness. Experiment results demonstrate that EPI-CGAN keeps a Correct Recognition Rate (CRR) over 98.5% on datasets that vary in participant conditions, scale, and EEG acquisition protocols.

Recently, state-of-the-art face manipulation algorithms have made significant progresses to forge images and videos that are able to deceive human eyes or even detection algorithms, which brings new challenges to forgery detection. In particular, the performance of detection algorithms for forgery videos is not as perfect as that for forgery images; furthermore, the difficulty of detection increases dramatically for low-quality videos. To address this challenge, we propose a novel dual stream architecture, referred to as FST-Net, for jointly mining forged features in the frequency, spatial and temporal domains. Specifically, we extract the spectral information of different frequency bands to expose intra-frame artifacts, and use the separable 3D CNN (S3D) to extract the spatio-temporal features among video frame groups. Moreover, to make the model focus on the tampered area, we add an attention layer to both backbone networks. Comprehensive experiments show that our model outperforms existing methods in video detection on challenging FaceForensics++ datasets, especially on low-quality video datasets.

Texture recognition is a challenging visual task since various primitives along with their arrangements can be recognized from a same texture image when perceiving with different contexts. Some recent work building on CNNs exploits orderless aggregating to provide invariance to spatial arrangements. However, these methods ignore the inherent structural property of textures, which is a critical cue for distinguishing and describing texture images in the wild. To address this problem, we propose a novel Deep Structure-Revealed Network (DSR-Net) that leverages spatial dependency among the captured primitives as structural representation for texture recognition. Specifically, a primitive capturing module (PCM) is devised to generate multiple primitives from eight directional spatial contexts, in which deep features are firstly extracted under the constrains of direction map and then encoded based on the similarities of neighborhood. Next, these primitives are associated with a dependence learning module (DLM) to generate structural representation, in which a two-way collaborative relationship strategy is introduced to perceive the spatial dependencies among multiple primitives. At last, the structure-revealed texture representations are integrated with spatial ordered information to achieve real-world texture recognition. Evaluation on the five most challenging texture recognition datasets has demonstrated the superiority of the proposed model against state-of-the-art methods. The structure-revealed performances of DSR-Net are further verified on some extensive experiments, including fine-grained classification and semantic segmentation.

In this paper, we aim to construct a spotting framework automatically. It is still a great challenge to spot micro-expression(ME) intervals accurately due to short duration, low intensity, and shaking. Under the uncontrolled condition, the transformation is caused by head shaking. In order to remove the global movement caused by head shaking, we propose a simple yet effective method to disentangle local movement vector from the global optical flow field by the estimation of mean optical flow in the nose region. After preprocessing, we extract the completed specific pattern(SP) of ME in each region of interest(ROI). The pattern consists of two sub-patterns: magnitude and angle. However, influenced by frame rate and different intensities of micro- and macro- expressions, we propose to use a multi-scale filter to improve the ability to spot both micro- and macro-expressions. The spotting result performed on $CAS(ME)^{2}$ and SAMM shows that our proposed method outperforms the basline method.

This paper proposes a novel collaborative attention mechanism, to fully utilize the mutually reinforcing relationship among the knowledge graph representation learning procedure (i.e., structure representation) and textual relation representation learning procedure (i.e., text representation). Based on this collaborative attention mechanism, a text-enhanced knowledge graph (KG) representation model is proposed, which could utilize textual information to enhance the knowledge representations and make the multi-direction signals to be fully integrated to learn more accurate textual representations for further improving structure representation and vice versa. Experimental results demonstrate the efficiency of the proposed model on both link prediction task and triple classification task.

Generalized zero-shot learning (GZSL) is a challenging class of vision and knowledge transfer problems in which both seen and unseen classes appear during testing. Existing GZSL approaches either suffer from semantic loss and discard discriminative information at the embedding stage, or cannot guarantee the visual-semantic interactions. To address these limitations, we propose a Dual Adversarial Semantics-Consistent Network (referred to as DASCN), which learns both primal and dual Generative Adversarial Networks (GANs) in a unified framework for GZSL. In DASCN, the primal GAN learns to synthesize inter-class discriminative and semantics-preserving visual features from both the semantic representations of seen/unseen classes and the ones reconstructed by the dual GAN. The dual GAN enforces the synthetic visual features to represent prior semantic knowledge well via semantics-consistent adversarial learning. To the best of our knowledge, this is the first work that employs a novel dual-GAN mechanism for GZSL. Extensive experiments show that our approach achieves significant improvements over the state-of-the-art approaches.

Real-time traffic volume inference is key to an intelligent city. It is a challenging task because accurate traffic volumes on the roads can only be measured at certain locations where sensors are installed. Moreover, the traffic evolves over time due to the influences of weather, events, holidays, etc. Existing solutions to the traffic volume inference problem often rely on dense GPS trajectories, which inevitably fail to account for the vehicles which carry no GPS devices or have them turned off. Consequently, the results are biased to taxicabs because they are almost always online for GPS tracking. In this paper, we propose a novel framework for the citywide traffic volume inference using both dense GPS trajectories and incomplete trajectories captured by camera surveillance systems. Our approach employs a high-fidelity traffic simulator and deep reinforcement learning to recover full vehicle movements from the incomplete trajectories. In order to jointly model the recovered trajectories and dense GPS trajectories, we construct spatiotemporal graphs and use multi-view graph embedding to encode the multi-hop correlations between road segments into real-valued vectors. Finally, we infer the citywide traffic volumes by propagating the traffic values of monitored road segments to the unmonitored ones through masked pairwise similarities. Extensive experiments with two big regions in a provincial capital city in China verify the effectiveness of our approach.

The fast pace of global urbanization is drastically changing the population distributions over the world, which leads to significant changes in geographical population densities. Such changes in turn alter the underlying geographical power demand over time, and drive power substations to become over-supplied (demand z capacity) or under-supplied (demand ≈ capacity). In this paper, we make the first attempt to investigate the problem of power substation-user assignment by analyzing large-scale power grid data. We develop a Scalable Power User Assignment (SPUA) framework, that takes large-scale spatial power user/substation distribution data and temporal user power consumption data as input, and control the assignments between users and substations, in a manner that minimizes the maximum substation utilization among all substations. To evaluate the performance of our SPUA framework, we conduct evaluations on real power consumption data and user/substation location data collected from a northwestern province in China for 35 days in 2015. The evaluation results demonstrate that our SPUA framework can achieve a 20-65 percent reduction on the maximum substation utilization, and 2 to 3.7 times reduction on total transmission loss over other baseline methods.

In large-scale distributed file systems, efficient metadata operations are critical since most file operations have to interact with metadata servers first. In existing distributed hash table (DHT) based metadata management systems, the lookup service could be a performance bottleneck due to its significant CPU overhead. Our investigations showed that the lookup service could reduce system throughput by up to 70 percent, and increase system latency by a factor of up to 8 compared to ideal scenarios. In this paper, we present MetaFlow, a scalable metadata lookup service utilizing software-defined networking (SDN) techniques to distribute lookup workload over network components. MetaFlow tackles the lookup bottleneck problem by leveraging B-tree, which is constructed over the physical topology, to manage flow tables for SDN-enabled switches. Therefore, metadata requests can be forwarded to appropriate servers using only switches. Extensive performance evaluations in both simulations and testbed showed that MetaFlow increases system throughput by a factor of up to 3.2, and reduce system latency by a factor of up to 5 compared to DHT-based systems. We also deployed MetaFlow in a distributed file system, and demonstrated significant performance improvement.

The fast pace of global urbanization is drastically changing the population distributions over the world, which leads to significant changes in geographical population densities. Such changes in turn alter the underlying geographical power demand over time, and drive power substations to become over-supplied (demand << capacity) or under-supplied (demand ≈ capacity). In this paper, we make the first attempt to investigate the problem of power substation-user assignment by analyzing large-scale power grid data. We develop a Scalable Power User Assignment (SPUA) framework, that takes large-scale spatial power user/substation distribution data and temporal user power consumption data as input, and assigns users to substations, in a manner that minimizes the maximum substation utilization among all substations. To evaluate the performance of our SPUA framework, we conduct evaluations on real power consumption data and user/substation location data collected from a province in China for 35 days in 2015. The evaluation results demonstrate that our SPUA framework can achieve a 20%--65% reduction on the maximum substation utilization, and 2 to 3.7 times reduction on total transmission loss over other baseline methods.

Cloud service providers are building out geographically distributed networks of datacenters around the world. It is customary for cloud service providers to distribute their data replicas at multiple geographic locations to mitigate user latency and to increase service availability. In this paper, we treat the content distributed from one datacenter to multiple datacenters as a multicast session. We investigate the problem of maximizing the capacity utilization of inter-datacenter networks while maintaining fairness among multiple multicast sessions. A bandwidth allocation algorithm based on max-min fairness is developed, named Blossom. Blossom leverages a fully polynomial time approximation scheme to accelerate the bandwidth allocation, while achieving an approximation that is (1- ϵ)-optimal. Through trace-driven simulation, we show that our approach is substantially more efficient than prior work.

Deploying caches in mobile networks, especially in the radio access network (RAN) is regarded as a promising way to improve mobile user experiences and alleviate the increasing pressure of traffic growth. However, the characteristics of mobile traffic and the performance of RAN caching still remains unclear. In this paper, we extensively analyze the traffic characteristics, the content popularity and the cache performance using a unique dataset collected from a commercial LTE network of China Mobile, from the perspective of mobile access network. The dataset spans nearly a week and consists of a collection of approximately 62.1 millions HTTP sessions, generated by more than 3200 users distributed across three base stations. Based on this realistic dataset, we observe that HTTP traffic can be reduced by 24.4% on average and the hit ratio can reach up to 42.2%, using 100GB cache size. The implications on some fundamental design issues of practical RAN caching systems, including reasonable size of RAN cache, suitable locations of cache deployment and potential benefits of collaborative RAN caching, are further presented. We believe our findings will shed light on practical RAN caching system design.

In recently proposed information centric networks (ICN), a user issues “interest” packets to retrieve contents from network by names. Once fetched from origin servers, “data” packets are replicated and cached in all routers along routing and forwarding paths, thus allowing further interests from other users to be fulfilled quickly. However, the way ICN caching and interest fulfillment work poses a great privacy risk: the time difference between responses for an interest of cached and uncached content can be used as an indicator to infer whether or not a near-by user has previously requested the same content as that requested by an adversary. This work introduces the extent to which the problem is applicable in ICN and provides several solutions that try to strike a balance between cost and benefits, and raise the bar for an adversary to apply such attack.

In content-centric networks, it is challenging how to optimally provision in-network storage to cache contents, to balance the tradeoffs between the network performance and the provisioning cost. To address this problem, we first propose a holistic model for intradomain networks to characterize the network performance of routing contents to clients and the network cost incurred by globally coordinating the in-network storage capability. We then derive the optimal strategy for provisioning the storage capability that optimizes the overall network performance and cost, and analyze the performance gains via numerical evaluations on real network topologies. Our results reveal interesting phenomena; for instance, different ranges of the Zipf exponent can lead to opposite optimal strategies, and the tradeoffs between the network performance and the provisioning cost have great impacts on the stability of the optimal strategy. We also demonstrate that the optimal strategy can achieve significant gain on both the load reduction at origin servers and the improvement on the routing performance. Moreover, given an optimal coordination level ℓ*, we design a routing-aware content placement (RACP) algorithm that runs on a centralized server. The algorithm computes and assigns contents to each CCN router to store, which can minimize the overall routing cost, e.g., transmission delay or hop counts, to deliver contents to clients. By conducting extensive simulations using a large-scale trace dataset collected from a commercial 3G network in China, our results demonstrate that our caching scheme can achieve 4% to 22% latency reduction on average over the state-of-the-art caching mechanisms.

Virtual machine migration has been touted as one of the crucial technologies in improving data center efficiency, such as reducing energy cost and maintaining load balance. However, traditional approaches could not avoid the service interruption completely. Moreover, they often result in longer delay and are prone to failures. In this paper, we leverage the emerging named data networking (NDN) to design an efficient and robust protocol to support seamless virtual machine migration in cloud data center. Specifically, virtual machines (VMs) are named with the services they provide. Request routing is based on service names instead of IP addresses that are normally bounded with physical machines. As such, services would not be interrupted when migrating supported VMs to different physical machines. We further analyze the performance of our proposed NDN-based VM migration protocol, and optimize its performance via a load balancing algorithm. Our extensive evaluations verify the effectiveness and the efficiency of our approach and demonstrate that it is interruption-free.

Named data networking architectures have emerged and attracted significant attention in recent years. Due to the name space being many orders of magnitude larger and more complex than IP address space, such architectures pose many significant challenges on design of routers and routing algorithms. IP forwarding algorithms are no longer applicable and cannot reach the satisfactory performance. In this paper, we ask ourselves the following question: how fast can the longest prefix name based lookup be? This is a key question to answer in order for named data networking architectures to be potentially successful. As an attempt to answer this question, we present a framework of fast longest prefix name lookup, based on a name space reduction scheme we proposed for named data networking. We also investigate feasible implementations on both CPU and GPU platforms. We demonstrate that we can achieve name lookup throughput as high as more than 37 and 70 million packets per second, respectively, on off-theshelf general-purpose CPU and GPU platforms. These results suggest that achieving high-performance lookup in named data networking architectures is not only feasible but also practical.

In recently proposed information centric networks (ICN), a user issues "interest" packets to retrieve contents from network by names. Once fetched from origin servers, "data" packets are replicated and cached in all routers along routing and forwarding paths, thus allowing further interests by other users to be fulfilled quickly. However, the way ICN caching works poses a great privacy risk: the time difference between responses for an interest of cached and uncached content can be used as an indicator to infer whether or not a near-by user has previously requested the same content as that requested by an adversary. This work introduces the extent to which the problem is applicable in ICN and provides several solutions that try to strike a balance between their cost and benefits, and raise the bar for the adversary to apply such attack.

Name-based routing in content-centric networks leads to a paradigm shift from a host-to-host to a host-to-content communication model. However, name-based routing faces many challenges. In this paper, we mainly focus on the scalability challenge and address it via a hybrid, reactive routing approach, where we augment the content-centric networking design with an Infrastructure Information Base (IIB) to reap the benefits of both name-based and infrastructure-based routing. Our preliminary evaluations demonstrate that the hybrid reactive routing approach can significantly improve the network scalability by reducing the number of broadcast packets by 2-3 orders of magnitude while maintaining the same level of user experiences.

In-network content storage has become an inherent capability of routers in the content-centric networking architecture. This raises new challenges in utilizing and provisioning the in-network caching capability, namely, how to optimally provision individual routers' storage to cache contents, so as to balance the trade-offs between the network performance and the provisioning cost. To address this problem, we first propose a holistic model to characterize the network performance of routing contents to clients and the network cost incurred by globally coordinating the in-network storage capability. We then derive the optimal strategy for provisioning the storage capability that optimizes the overall network performance and cost, and analyze the performance gains via numerical evaluations on real network topologies. Our results reveal interesting phenomena; for instance, different ranges of the Zipf exponent can lead to opposite optimal strategies, and the trade-offs between the network performance and the provisioning cost have great impacts on the stability of the optimal strategy. We also demonstrate that the optimal strategy can achieve significant gain on both the load reduction at origin servers and the improvement on the routing performance.

In recent years, technical challenges are emerging on how to efficiently distribute the rapid growing user-generated contents (UGCs) with long-tailed nature. To address this issue, we have previously proposed cloud centric media network (CCMN) for cost-efficient UGCs delivery. In this paper, we further study the content placement problem in CCMN. Our objective is to minimize the monetary cost incurred by using cloud resources to orchestrate an elastic and global content delivery network (CDN) service. In particular, this objective is achieved via a two-step method. First, for a single content, we map it into a k-center problem, and find a logarithmic relationship between the mean hop distance from users to contents, and the reciprocal of replica number. Second, for multiple contents, we formulate a convex optimization with storage and bandwidth capacity constraints, which can be solved by our proposed algorithm. Finally, we verify the algorithm based on real-world traces collected from a popular video website in China. Our numerical results suggest that, the optimal number of replica for each content follows a power law in respect to its popularity, under feasible storage and bandwidth constraints, in a set of deployed backbone networks.

It has been a significant challenge for network carriers to deploy and provision a large number of Customer-Premises Equipment (CPE) devices located at subscribers' premises and connected to a carrier's network infrastructure. In this paper, we make a first systematic attempt to fundamentally re-shape the access networks into a software defined networking architecture by virtualizing the network functionality of residential gateways (vRGW). Our approach can be generalized to other CPE such as set-top boxes. Our analysis suggests that vRGW can achieve significant economic benefits ranging from up to 90% reduction on the call center cost and up to 46% reduction on the product return cost.

Nowadays, TV experience has been transformed from the traditional "laid-back" video watching experience to a "lean-forward" social and multi-screen experience. In this demo, we design and develop a multi-screen cloud social TV system in response to this trend. Our system is built upon two enabling technologies, including a cloud based back-end infrastructure and a multi-screen front-end application. We demonstrate two key features of our system based on real user scenarios, including a living-room video watching experience with remote viewers, and the video teleportation as an enhanced multi-screen experience.

The IPv6 transition has been an ongoing process throughout the world due to the exhaustion of the IPv4 address space. However, this transition leads to costly end-to-end network upgrades and poses new challenges of managing a large number of devices with a variety of transitioning protocols. Recognizing these difficulties, we propose an software defined approach to unifying the deployment of IPv6 in a cost-effective, flexible manner. Our deployment and experiments demonstrate significant benefits of this approach, including low complexity, low cost and high flexibility of adopting different existing transition mechanisms.

Longest-prefix name-based forwarding in Content-Centric Networking poses many significant challenges on design of routers and algorithms, because the name space is many orders of magnitude larger and significantly more complex than the IP address space. IP forwarding algorithms are no longer applicable and cannot reach the satisfactory performance. In this paper, we present a framework of fast longest-name-prefix lookup, based on a name space reduction scheme we proposed for content-centric networking. We implement the algorithm using fat tree and extensible hybrid data structures to speed up name lookup. Our evaluations demonstrate that we can achieve name lookup throughput as high as more than 37 Mpps on off-the-shelf general-purpose CPU platforms.

In information centric network (ICN), contents are fetched by their names from caches deployed in the network or from origin servers. Once the contents are fetched from the origin server, it is replicated and cached in all routers along the routing and forwarding paths from the user that issues the interest to the origin server, thus allowing further "interests" by other users to be fulfilled quickly. However, the way ICN caching and interest fulfillment work pose a great privacy risk; the time difference between response for interest of cached and uncached contents can be used as an indicator to infer whether or not a near-by user previously requested the same contents requested by the adversary. This work introduces the extent to which the problem is applicable in ICN and provides several solutions to address it.

There has been an increasingly popular trend that content caching becomes an inherent underlay capability in network routers. This poses new challenges, and we believe that if not well provisioned, such capability may even degrade network performance. Among the challenges, we are interested in how the in-network caching capability should be best provisioned to improve the overall network performance. To address this challenge, we propose a collaborative caching scheme guided by traffic engineering (TECC) for the emerging content-centric networks. In particular, we decouple collaborative in-network caching and traffic engineering via the primal-dual decomposition. Our evaluations, although preliminary, suggest that the improvement is as much as 80% compared to collaborative caching without guidance of traffic engineering.

We study a streaming network application-video transcoding to be executed on a multi-core server. It is important for the scheduler to minimize the total processing time and preserve good video quality in an energy-efficient manner. However, the performance of existing scheduling schemes is largely limited by ineffective use of the multi-core architecture characteristic and undifferentiated transcoding cost in terms of energy consumption. In this paper, we identify three key factors that collectively play important roles in affecting transcoding performance: memory access (M), core/cache topology (C) and transcoding format cost (C), or MC2for short. Based on MC2, we propose E-AHRW, an Energy-efficient Adaptive Highest Random Weight hash scheduler by extending the HRW scheduler proposed for packet scheduling on a homogeneous multiprocessor. E-AHRW achieves stream locality and load balancing at both stream and packet (frame) level by adaptively adjusting the hashing decision according to real-time weighted queue length of each processing unit (PU). Based on E-AHRW, we also design, implement and evaluate a hash-tree scheduler to further reduce the computation cost and achieve more effective load balancing on multi-core architectures. Through implementation on an Intel Xeon server and evaluations on realistic workload, we demonstrate that E-AHRW improves throughput, energy efficiency and video quality due to better load balancing, lower L2 cache miss rate and negligible scheduling overhead.

Peer-to-Peer (P2P) applications have been consuming an increasingly significant fraction of Internet bandwidth. They are becoming a financial burden to Internet Service Providers (ISPs), creating hot spots in the Internet, and causing potential performance degradation to other applications. As a result, there has been increasing tensions between P2P applications and network service providers. In this paper, we propose a framework called SODA for P2P applications to be self-adaptive and optimize their demands in order to more efficiently utilize network resources. Through preliminary experiments using two representative P2P applications, we demonstrate that SODA can effectively reduce bandwidth consumption by adapting the demands among peers.

Inference of the routing topology and link performance from a node to a set of other nodes is an important component of network monitoring and application design. In this paper we propose a general framework for designing topology inference algorithms based on additive metrics. Our framework allows the integration of both end-to-end packet probing measurements and traceroute type measurements. Based on this framework we design several computationally efficient topology inference algorithms. In particular, we propose a novel sequential topology inference algorithm to address the probing scalability problem and handle dynamic node joining and leaving. We provide sufficient conditions for the correctness of our algorithms and derive lower bounds on the probability of correct topology inference. We conduct Internet experiments to evaluate and demonstrate the effectiveness of our algorithms.

Traffic engineering plays a critical role in determining the performance and reliability of a network. A major challenge in traffic engineering is how to cope with dynamic and unpredictable changes in traffic demand. In this paper, we propose COPE, a class of traffic engineering algorithms that optimize for the expected scenarios while providing a worst-case guarantee for unexpected scenarios. Using extensive evaluations based on real topologies and traffic traces, we show that COPE can achieve efficient resource utilization and avoid network congestion in a wide variety of scenarios.

We present a general model of interdomain route selection to study interdomain traffic engineering. In this model, the routing of multiple destinations can be coordinated. Thus the model can capture general traffic engineering behaviors such as load balancing and link capacity constraints. We first identify potential routing instability and inefficiency of interdomain traffic engineering. We then derive a sufficient condition to guarantee convergence. We also show that the constraints on local policies imposed by business considerations in the Internet can guarantee stability without global coordination. Using realistic Internet topology, we evaluate the extent to which routing instability of interdomain traffic engineering can happen when the constraints are violated.

The recent discovery of instability caused by the interaction of local routing policies of multiple ASes has led to extensive research on the subject. However, previous studies analyze stability under a specific route selection algorithm. In this paper, instead of studying a specific route selection algorithm, we study a general class of route selection algorithms which we call rational route selection algorithms. We present a sufficient condition to guarantee routing convergence in a heterogeneous network where each AS runs any rational route selection algorithm. Applying our general results, we study the potential instability of a network where the preference of an AS depends on not only its egress routes to the destinations but also its inbound traffic patterns (i.e., the distribution of incoming traffic from its neighbors). We show that there exist networks which will have persistent route oscillations even when the ASes strictly follow the constraints imposed by business considerations, and adopt any rational route selection algorithms.

Multihoming is a popular method used by large enterprises and stub ISPs to connect to the Internet to reduce cost and improve performance. Recently researchers have studied the potential benefits of multihoming and proposed protocols and algorithms to realize these benefits. They focus on how to dynamically select which ISPs to use for forwarding and receiving packets, and assume that the set of subscribed ISPs is given a priori. In practice, a user often has the freedom to choose which subset of ISPs among all available ISPs to subscribe to. We call the problem of how to choose the optimal set of ISPs the ISP subscription problem. In this paper, We design a dynamic programming algorithm to solve the ISP subscription problem optimally. We also design a more efficient algorithm for a large class of common pricing functions. Using real traffic traces and realistic pricing data, we show that our algorithm reduces users' cost. Next we study how ISPs respond to users' optimal ISP subscription by adjusting their pricing strategies. We call this problem the ISP pricing problem. Using a realistic charging model, we formulate the problem as a non-cooperative game. We first prove that if cost is the only criterion used by a user to determine which subset of ISPs to subscribe to, at any equilibrium all ISPs receive zero revenue. We then study a more practical formulation in which different ISPs provide different levels of reliability and users choose ISPs to both improve reliability and reduce cost. We analyze this problem and show that at any equilibrium an ISP's revenue is positive and determined by its reliability.

Recently we have seen an emergent trend of self adaptive routing in both Internet and wireless ad hoc networks. Although there are previous methods for computing the traffic equilibria of self adaptive routing (e.g., selfish routing), these methods use computationally demanding algorithms and require that a precise analytical model of the network be given. Also, it remains an open question how to design an adaptive routing scheme which ensures convergence to traffic equilibria in practice. In this paper we propose a simple, efficient, distributed probabilistic routing scheme for self adaptive routing in dynamic, realistic environments. Using both analysis and extensive simulations, we show that our scheme can converge to the desired traffic equilibrium (either user-optimal or network-optimal) very quickly. We find that user-optimal routing can achieve very close to optimal average latency in dynamic environments, but such performance often comes at the cost of seriously overloading certain links. To avoid link overloads, we improve adaptive routing by optimizing average user latency and link utilization simultaneously. Our evaluation shows that there is a trade-off between optimizing dual objectives, but the degradation in average latency is only marginal for typical link utilization requirements.

Multihoming is often used by large enterprises and stub ISPs to connect to the Internet. In this paper, we design a series of novel smart routing algorithms to optimize cost and performance for multihomed users. We evaluate our algorithms through both analysis and extensive simulations based on realistic charging models, traffic demands, performance data, and network topologies. Our results suggest that these algorithms are very effective in minimizing cost and at the same time improving performance. We further examine the equilibrium performance of smart routing in a global setting and show that a smart routing user can improve its performance without adversely affecting other users.