INFOCOM 2017 论文列表

The separated control and data planes in software-defined networking (SDN) with high programmability introduce a more flexible way to manage and control network traffic. However, SDN will experience long packet delay and high packet loss rate when the communication link between two planes is jammed by SDN-aimed DoS attacks with massive table-miss packets. In this paper, we propose FloodDefender, an efficient and protocol-independent defense framework for SDN/OpenFlow networks to mitigate DoS attacks. It stands between the controller platform and other controller apps, and can protect both the data and control plane resources by leveraging three new techniques: table-miss engineering to prevent the communication bandwidth from being exhausted; packet filter to identify attack traffic and save computational resources of the control plane; and flow rule management to eliminate most of useless flow entries in the switch flow table. All designs of FloodDefender conform to the OpenFlow policy, requiring no additional devices. We implement a prototype of FloodDefender and evaluate its performance in both software and hardware environments. Experimental results show that FloodDefender can efficiently mitigate the SDN-aimed DoS attacks, incurring less than 0.5% CPU computation to handle attack traffic, only 18ms packet delay and 5% packet loss rate under attacks.

We consider a model of two interdependent networks, where every node in one network depends on one or more supply nodes in the other network and a node fails if it loses all of its supply nodes. We develop algorithms to compute the failure probability of a path, and obtain the most reliable path between a pair of nodes in a network, under the condition that each supply node fails independently with a given probability. Our work generalizes the classical shared risk group model, by considering multiple risks associated with a node and letting a node fail if all the risks occur. Moreover, we study the diverse routing problem by considering two paths between a pair of nodes. We define two paths to be d-failure resilient if at least one path survives after removing d or fewer supply nodes, which generalizes the concept of disjoint paths in a single network, and risk-disjoint paths in a classical shared risk group model. We compute the probability that both paths fail, and develop algorithms to compute the most reliable pair of paths.

With social media platform such as Twitter becoming the de facto destination for users' views and opinions, it is of great importance to forecast an information outbreak. In Twitter, tweets are often annotated with hashtags to help its users to quickly extract their contents. The existing approaches for modeling the dynamics of tweet-messages are usually limited to individual or simple aggregates of tweets rather than the underlying hashtags. In this paper, we develop, STRM, a novel point process driven model that considers the effect of cross-tweet impact in hashtag popularity. STRM, by assuming hashtag to be a heterogeneous collection of tweet-chains. Through extensive experimentation, we find that our algorithm - STRM, shows consistent performance boosts with six diverse real datasets against several strong baselines. Moreover surprisingly, it also offers significant accuracy gains in popularity-prediction for individual tweets as compared with the existing paradigms.

Adaptive bitrate (ABR) streaming is the de facto solution in online video services to cope with heterogeneous devices and varying network connections. However, this solution is computation intensive, demanding a large number of servers for encoding videos. Moreover, due to the time-varying nature of video generation, intelligent strategies are required in order to determine the right amount of resources for encoding. The situation is further complicated by the fact that, the two types of co-existing video content, live content and Video-on-Demand (VoD) content, have different QoS requirements for encoding. These observations posit daunting challenges for meeting the heterogeneous QoS requirements with a minimum computing capacity. This paper proposes the QoS-differentiated low-cost video encoding (QDLCoding) scheme to address these challenges. We develop a framework for scheduling the encoding workloads of the two types of videos with statistical QoS guarantees. Each type of videos is specified with a QoS criterion and a QoS loss bound. The objective is to provision the minimum amount of resources while keeping the QoS loss probabilities within the prescribed bounds. We design an online algorithm that can determine the minimum required capacity by learning content arrival distributions. The experiment results demonstrate that our method can greatly reduce the required capacity for encoding online videos while controlling the likelihood of QoS loss precisely.

Cloud computing provides a cost-effective computing platform for big data workflows where moldable parallel computing models such as MapReduce are widely applied to meet stringent performance requirements. The granularity of task partitioning in each moldable job has a significant impact on workflow completion time and financial cost. We investigate the properties of moldable jobs and design a big-data workflow mapping model, based on which, we formulate a workflow mapping problem to minimize workflow makespan under a budget constraint in public clouds. We show this problem to be strongly NP-complete and design i) a fully polynomial-time approximation scheme (FPTAS) for a special case with a pipeline-structured workflow executed on virtual machines in a single class, and ii) a heuristic for a generalized problem with an arbitrary directed acyclic graph-structured workflow executed on virtual machines in multiple classes. The performance superiority of the proposed solution is illustrated by extensive simulation-based results in Hadoop/YARN in comparison with existing workflow mapping models and algorithms.

We introduce structural transformations that allow simplifying a given network while preserving its original “bandwidth” and “routing” capabilities, transparently to specific allocations. We minimize a certain objective such as the aggregate capacity of network links, number of nodes, or number of links, in such a way that all the bandwidth that could be routed in the original network can also be routed in the reduced one. This improves cost-efficiency for both inter- and intra-datacenter connections and simplifies network management. We also identify a fundamental tradeoff between extra added capacity and simplicity of representation for a given network. Our analytic results are supported by extensive simulation results on hundreds of real network topologies. One result is that by adding 10-30% extra capacity to evaluated real-world networks one can simplify them down to a star topology with a single switch, while all routing and bandwidth allocation decisions on the simplified topology can be mapped back to the original network. This is an important step towards simplifying network management via a reduced virtualized network infrastructure.

In edge-cloud computing, a set of edge servers are deployed near the mobile devices such that these devices can offload jobs to the servers with low latency. One fundamental and critical problem in edge-cloud systems is how to dispatch and schedule the jobs so that the job response time (defined as the interval between the release of a job and the arrival of the computation result at its device) is minimized. In this paper, we propose a general model for this problem, where the jobs are generated in arbitrary order and times at the mobile devices and offloaded to servers with both upload and download delays. Our goal is to minimize the total weighted response time over all the jobs. The weight is set based on how latency sensitive the job is. We derive the first online job dispatching and scheduling algorithm in edge-clouds, called OnDisc, which is scalable in the speed augmentation model; that is, OnDisc is (1 + ε)-speed O(1/ε)-competitive for any constant ε ϵ (0,1). Moreover, OnDisc can be easily implemented in distributed systems. Extensive simulations on a real-world data-trace from Google show that OnDisc can reduce the total weighted response time dramatically compared with heuristic algorithms.

In order to cope with the exponential growth of mobile traffic, mobile operators need to access more spectrum resources. LTE in unlicensed spectrum (LTE-U) has been proposed to extend the usual operation of LTE in licensed spectrum to cover also unlicensed spectrum. However, this extension poses significant challenges especially regarding the coexistence between LTE-U and legacy systems like Wi-Fi. In case of LTE-U adopts Time-Division Multiplexing (TDM) schemes to share the spectrum with Wi-Fi, we expect performance degradations of Wi-Fi networks. In this paper, we quantify the impact of TDM schemes on Wi-Fi performance in a coexistence scenario. We provide detailed analytical models using two different random walk approaches to compute the probability of collision faced by Wi-Fi stations and their throughput performance. Besides, we derive the performance results using an exponential approximation which shows its insufficiency to capture the exact behavior. We implement the coexistence in the NS3 simulator and we show that the models estimate accurately the collision probability and the throughput experienced by Wi-Fi. The models are then used to study and compare different coexistence schemes showing for instance that the Wi-Fi frame size has a non-negligible impact on the performance of Wi-Fi users.

System-level timing inconsistency and wireless communication delays are the main uncertainties of existing synchronization mechanisms for wireless sensor networks. Existing solutions mainly rely on timestamp exchanges to estimate clock offset and skew, which results in frequent synchronization, high overhead, and high energy consumption to maintain a well-synchronized network. This paper introduces a novel clock syntonization approach to estimate the differences between clock frequencies of network nodes without the need for timestamp exchanges. The carrier frequency offset (CFO) assisted syntonization (CFOSynt) utilizes the carrier information obtained from wireless packet transmission for clock skew compensation. The key idea of CFOSynt is that, in any wireless communication system, where carrier modulation is employed, carrier frequency delivers information about the transmitter RF clock. Consequently, clock frequency offset between a pair of sensor nodes will result in a carrier frequency offset detected by the receiver node. By leveraging the CFO information, CFOSynt can estimate the system clock skew based on digital counter theory. Extensive experiments and numerical analysis have been demonstrated to evaluate clock skew estimation.

Live migration of virtual machines (VM) is one of the key characteristics of virtualization for load balancing, system maintenance, power management, etc., in data centers or clusters. In order to reduce the data transferred and shorten the migration time, the compression techniques have been widely used to accelerate VM migration. However, different compression approaches have different compression ratios and speeds. Because there is a trade-off between compression and transmission, the migration performance improvements obtained from different compression approaches are differentiated, and the improvements vary with the network bandwidth. Besides, the compression window sizes used in most compression algorithms are typically much larger than a single page size, so the traditional single page compression loses some potential compression benefits. In this paper, we design and implement a Bandwidth-Aware Compression (BAC) scheme for VM migration. BAC chooses suitable compression approach according to the network bandwidth available for the migration process, and employs multi-page compression. These features make BAC obtain more migration performance improvements from compression. Experiments under various network scenarios demonstrate that, compared with conventional compression approaches, BAC shortens the total migration time while achieving comparable performance for the total data transferred and the downtime.

Cloud computing has emerged as a powerful and elastic platform for internet service hosting, yet it also draws concerns of the unpredictable performance of cloud-based services due to network congestion. To offer predictable performance, the virtual cluster abstraction of cloud services has been proposed, which enables allocation and performance isolation regarding both computing resources and network bandwidth in a simplified virtual network model. One issue arisen in virtual cluster allocation is the survivability of tenant services against physical failures. Existing works have studied virtual cluster backup provisioning with fixed primary embeddings, but have not considered the impact of primary embeddings on backup resource consumption. To address this issue, in this paper we study how to embed virtual clusters survivably in the cloud data center, by jointly optimizing primary and backup embeddings of the virtual clusters. We formally define the survivable virtual cluster embedding problem. We then propose a novel algorithm, which computes the most resource-efficient embedding given a tenant request. Since the optimal algorithm has high time complexity, we further propose a faster heuristic algorithm, which is several orders faster than the optimal solution, yet able to achieve similar performance. Besides theoretical analysis, we evaluate our algorithms via extensive simulations.

In the smart grid, computer networks (i.e., the cyber domain) are built upon physical infrastructures (i.e., the physical domain) to facilitate advanced functionalities that were considered not possible in legacy systems. It is envisioned that such a cyber-physical paradigm enables intelligent, collaborative controls to prevent faults from propagating along large-scale infrastructures, which is a primary cause for massive blackouts (e.g., Northeast blackout of 2003). Despite this promising vision, how effective cyber and physical interactions are against fault propagation is not yet fully investigated. In this paper, we use analysis and system-level simulations to characterize such interactions during load shedding, which is a process to stop fault propagation by shedding a computed amount of loads based on collaborative communication. Specifically, we model faults happening in the physical domain as a counting process, with each count triggering a load shedding action on the fly in the cyber domain. We show that although global load shedding design is considered optimal by globally coordinating shedding actions in power engineering, its induced failure probability (defined as the one that at least a given number of power lines fail) is scalable to the delay performance and the system size in the cyber domain, thus less likely to stop fault propagation in large systems than local shedding design that sheds loads within a limited system scope. Our study demonstrates that a joint view on cyber and physical factors is essential for failure prevention design in the smart grid.

Network Function Virtualization (NFV) represents a new paradigm of network service provisioning. NFV providers acquire cloud resources, install virtual network functions (VNFs), assemble VNF service chains for customer usage, and dynamically scale VNF deployment against input traffic fluctuations. While existing literature on VNF scaling mostly adopts a reactive approach, we target a proactive approach that is more practical given the time overhead for VNF deployment. We aim to effectively estimate upcoming traffic rates and adjust VNF deployment a priori, for flow service quality assurance and resource cost minimization. We adapt online learning techniques for predicting future service chain workloads. We further combine the online learning method with a multi-timescale online optimization algorithm for VNF scaling, through minimization of the regret due to inaccurate demand prediction and minimization of the cost incurred by sub-optimal online decisions in a joint online optimization framework. The resulting proactive online VNF provisioning algorithm achieves a good performance guarantee, as shown by both theoretical analysis and simulation under realistic settings.

We address secure communications over energy-harvesting based OFDMA cooperative cognitive radio networks, where one primary user (PU) cooperates with several secondary users (SUs) in terms of both information transmission and energy harvesting. To improve spectrum utilization and ensure that SU transmitters can harvest as much energy as possible, we suppose that SUs are equipped with orthogonally dual-polarized antennas. Based on these setting-ups, we propose the polarization-enabled two-phase cooperative framework, where SU transmitters first apply power splitting technique to harvest energy from radio frequency signals radiated by the PU transmitter, and then use the harvested energy to concurrently transmit their own and the PU's data. Under the proposed framework, we develop the secure resource allocation schemes for the scenarios when SU receivers are untrusted users, implying that each SU receiver may overhear the PU's and the other SUs' confidential information. For this scenario, which has hardly been studied, we investigate the joint allocation of relays, subcarriers, power splitting ratios, and powers, with the objective to maximize the total secrecy rate of all SUs while guaranteeing the PU's minimum secrecy rate. Finally, we validate and evaluate our proposed cooperative framework and resource allocation schemes through numerical analyses.

Crowdsourced mobile video streaming enables nearby mobile video users to aggregate their network resources to improve the video streaming performance. However, users are often selfish and may not be willing to cooperate without proper incentives. Designing an incentive mechanism for such a scenario is challenging due to the users' asynchronous downloading behaviors as well as their private valuations for multi-bitrate encoded videos. In this work, we propose a multi-object multi-dimensional auction-based incentive framework, through which users can download multiple video segments with different bitrates for multiple nearby users (and themselves). Based on this incentive framework, we propose a Vickrey-score auction, which is the first multi-object multi-dimensional auction that achieves both truthfulness and efficiency. Simulations with real traces show that crowdsourced mobile streaming outperforms noncooperative streaming by 48.6% (on average) in terms of social welfare. We further implement our proposed auction mechanism in a demostration system, and show that the crowdsourced framework together with the auction mechanism can substantially increase mobile user's welfare and video service stability.

The integration of cloud computing and Internet of Things (loT) is quickly becoming the key enabler for the digital transformation of the healthcare industry by offering comprehensive improvements in patient engagements, productivity and risk mitigation. This paradigm shift, while bringing numerous benefits and new opportunities to healthcare organizations, has raised a lot of security and privacy concerns. In this paper, we present a reliable, searchable and privacy-preserving e-healthcare system, which takes advantage of emerging cloud storage and IoT infrastructure and enables healthcare service providers (HSPs) to realize remote patient monitoring in a secure and regulatory compliant manner. Our system is built upon a novel dynamic searchable symmetric encryption scheme with forward privacy and delegated verifiability for periodically generated healthcare data. While the forward privacy is achieved by maintaining an increasing counter for each keyword at an IoT gateway, the data owner delegated verifiability comes from the combination of the Bloom filter and aggregate message authentication code. Moreover, our system is able to support multiple HSPs through either data owner assistance or delegation. The detailed security analysis as well as the extensive simulations on a large data set with millions of records demonstrate the practical efficiency of the proposed system for real world healthcare applications.

Key agreement using physical layer properties of communication channels is a well studied problem. iJam is a physical layer key agreement protocol that achieves security by creating a “virtual” wiretap channel for the adversary through a subprotocol between the sender and the receiver that uses self-jamming by the receiver. The protocol was implemented and its security was shown through extensive experiments. The self-jamming subprotocol of iJam was later modelled as a wiretap channel and used for designing a secure message transmission protocol with provable security. We use the same wiretap model of the subprotocol to design secret key agreement protocols with provable security. We propose two protocols that use the wiretap channel once from Alice to Bob, and a protocol that uses two wiretap channels, one from Alice to Bob, and one in the opposite direction. We provide security proof and efficiency analysis for the protocols. The protocols effectively give physical layer security protocols that can be implemented and have provable security. We discuss our results and propose directions for future research.

This paper studies the workload offloading problem for fog computing networks in which a set of fog nodes can offload part or all the workload originally targeted to the cloud data centers to further improve the quality-of-experience (QoE) of users. We investigate two performance metrics for fog computing networks: users' QoE and fog nodes' power efficiency. We observe a fundamental tradeoff between these two metrics for fog computing networks. We then consider cooperative fog computing networks in which multiple fog nodes can help each other to jointly offload workload from cloud data centers. We propose a novel cooperation strategy referred to as offload forwarding, in which each fog node, instead of always relying on cloud data centers to process its unprocessed workload, can also forward part or all of its unprocessed workload to its neighboring fog nodes to further improve the QoE of its users. A distributed optimization algorithm based on distributed alternating direction method of multipliers (ADMM) via variable splitting is proposed to achieve the optimal workload allocation solution that maximizes users' QoE under the given power efficiency. We consider a fog computing platform that is supported by a wireless infrastructure as a case study to verify the performance of our proposed framework. Numerical results show that our proposed approach significantly improves the performance of fog computing networks.

Various topologies have been proposed in the context of high-performance computing and data center networking. Direct-connect topologies generally offer large capacity with high path diversity and are highly cost effective for general data center traffic patterns. However, the lack of simple yet efficient load balancing techniques for direct-connect fabrics has hindered these networks from gaining traction in data centers. This paper presents the design, implementation, and evaluation of Flicr, a light-weight host-based load balancing mechanism for direct-connect data centers. Flicr dynamically reroutes traffic through minimal and non-minimal routes to avoid congesting the fabric. This enables Flicr to efficiently minimize networking resource consumption while exploiting high path diversity in direct-connect fabrics to balance the network and gracefully handle link failures. Flicr requires only a simple kernel modification and is readily deployable in commodity data centers today. Our evaluations show that Flicr consistently outperforms other state-of-the-art load balancing designs, achieving 25-60% lower average flow completion time compared to adaptive routing. Flicr is also more robust against link failures and has 5-8 χ better performance relative to other schemes in the presence of link failures.