Showing papers on "Server published in 2017"
TL;DR: The Mirai botnet and its variants and imitators are a wake-up call to the industry to better secure Internet of Things devices or risk exposing the Internet infrastructure to increasingly disruptive distributed denial-of-service attacks.
Abstract: The Mirai botnet and its variants and imitators are a wake-up call to the industry to better secure Internet of Things devices or risk exposing the Internet infrastructure to increasingly disruptive distributed denial-of-service attacks.
1,391 citations
22 May 2017
TL;DR: This paper presents new and efficient protocols for privacy preserving machine learning for linear regression, logistic regression and neural network training using the stochastic gradient descent method, and implements the first privacy preserving system for training neural networks.
Abstract: Machine learning is widely used in practice to produce predictive models for applications such as image processing, speech and text recognition. These models are more accurate when trained on large amount of data collected from different sources. However, the massive data collection raises privacy concerns. In this paper, we present new and efficient protocols for privacy preserving machine learning for linear regression, logistic regression and neural network training using the stochastic gradient descent method. Our protocols fall in the two-server model where data owners distribute their private data among two non-colluding servers who train various models on the joint data using secure two-party computation (2PC). We develop new techniques to support secure arithmetic operations on shared decimal numbers, and propose MPC-friendly alternatives to non-linear functions such as sigmoid and softmax that are superior to prior work. We implement our system in C++. Our experiments validate that our protocols are several orders of magnitude faster than the state of the art implementations for privacy preserving linear and logistic regressions, and scale to millions of data samples with thousands of features. We also implement the first privacy preserving system for training neural networks.
1,164 citations
TL;DR: A real-time, context-aware collaboration framework that lies at the edge of the RAN, comprising MEC servers and mobile devices, and amalgamates the heterogeneous resources at theedge is envisions.
Abstract: MEC is an emerging paradigm that provides computing, storage, and networking resources within the edge of the mobile RAN. MEC servers are deployed on a generic computing platform within the RAN, and allow for delay-sensitive and context-aware applications to be executed in close proximity to end users. This paradigm alleviates the backhaul and core network and is crucial for enabling low-latency, high-bandwidth, and agile mobile services. This article envisions a real-time, context-aware collaboration framework that lies at the edge of the RAN, comprising MEC servers and mobile devices, and amalgamates the heterogeneous resources at the edge. Specifically, we introduce and study three representative use cases ranging from mobile edge orchestration, collaborative caching and processing, and multi-layer interference cancellation. We demonstrate the promising benefits of the proposed approaches in facilitating the evolution to 5G networks. Finally, we discuss the key technical challenges and open research issues that need to be addressed in order to efficiently integrate MEC into the 5G ecosystem.
700 citations
TL;DR: It is argued that blockchain’s decentralized nature is likely to result in a low susceptibility to manipulation and forgery by malicious participants, and could be superior to the current IoT ecosystem, which relies mainly on centralized cloud servers.
Abstract: This column evaluates blockchain’s roles in strengthening security in the Internet of Things (IoT). Key underlying mechanisms related to the blockchain–IoT security nexus are covered. From a security standpoint, the article highlights how blockchain-based solutions could be, in many aspects, superior to the current IoT ecosystem, which relies mainly on centralized cloud servers. Using practical applications and real-world examples, the article argues that blockchain’s decentralized nature is likely to result in a low susceptibility to manipulation and forgery by malicious participants. Special consideration is given to how blockchain-based identity and access management systems can address some of the key challenges associated with IoT security. The column provides a detailed analysis and description of blockchain’s roles in tracking the sources of insecurity in supply chains related to IoT devices. Using blockchain, it is also possible to contain an IoT security breach in a targeted way after it is discovered. The column also discusses and evaluates initiatives of organizations, interorganizational networks, and industries on the frontlines of blockchain.
659 citations
07 Aug 2017
TL;DR: The experience with QUIC is presented, an encrypted, multiplexed, and low-latency transport protocol designed from the ground up to improve transport performance for HTTPS traffic and to enable rapid deployment and continued evolution of transport mechanisms.
Abstract: We present our experience with QUIC, an encrypted, multiplexed, and low-latency transport protocol designed from the ground up to improve transport performance for HTTPS traffic and to enable rapid deployment and continued evolution of transport mechanisms. QUIC has been globally deployed at Google on thousands of servers and is used to serve traffic to a range of clients including a widely-used web browser (Chrome) and a popular mobile video streaming app (YouTube). We estimate that 7% of Internet traffic is now QUIC. We describe our motivations for developing a new transport, the principles that guided our design, the Internet-scale process that we used to perform iterative experiments on QUIC, performance improvements seen by our various services, and our experience deploying QUIC globally. We also share lessons about transport design and the Internet ecosystem that we learned from our deployment.
610 citations
TL;DR: This paper develops an online joint radio and computational resource management algorithm for multi-user MEC systems, with the objective of minimizing the long-term average weighted sum power consumption of the mobile devices and the MEC server, subject to a task buffer stability constraint.
Abstract: Mobile-edge computing (MEC) has recently emerged as a prominent technology to liberate mobile devices from computationally intensive workloads, by offloading them to the proximate MEC server. To make offloading effective, the radio and computational resources need to be dynamically managed, to cope with the time-varying computation demands and wireless fading channels. In this paper, we develop an online joint radio and computational resource management algorithm for multi-user MEC systems, with the objective of minimizing the long-term average weighted sum power consumption of the mobile devices and the MEC server, subject to a task buffer stability constraint. Specifically, at each time slot, the optimal CPU-cycle frequencies of the mobile devices are obtained in closed forms, and the optimal transmit power and bandwidth allocation for computation offloading are determined with the Gauss-Seidel method ; while for the MEC server, both the optimal frequencies of the CPU cores and the optimal MEC server scheduling decision are derived in closed forms. Besides, a delay-improved mechanism is proposed to reduce the execution delay. Rigorous performance analysis is conducted for the proposed algorithm and its delay-improved version, indicating that the weighted sum power consumption and execution delay obey an $\left [{O\left ({1 / V}\right), O\left ({V}\right) }\right ]$ tradeoff with $V$ as a control parameter. Simulation results are provided to validate the theoretical analysis and demonstrate the impacts of various parameters.
576 citations
Posted Content•
TL;DR: In this article, the authors present new and efficient protocols for privacy preserving machine learning for linear regression, logistic regression and neural network training using the stochastic gradient descent method, where data owners distribute their private data among two non-colluding servers who train various models on the joint data using secure two-party computation.
Abstract: Machine learning is widely used in practice to produce predictive models for applications such as image processing, speech and text recognition. These models are more accurate when trained on large amount of data collected from different sources. However, the massive data collection raises privacy concerns. In this paper, we present new and efficient protocols for privacy preserving machine learning for linear regression, logistic regression and neural network training using the stochastic gradient descent method. Our protocols fall in the two-server model where data owners distribute their private data among two non-colluding servers who train various models on the joint data using secure two-party computation (2PC). We develop new techniques to support secure arithmetic operations on shared decimal numbers, and propose MPC-friendly alternatives to non-linear functions such as sigmoid and softmax that are superior to prior work. We implement our system in C++. Our experiments validate that our protocols are several orders of magnitude faster than the state of the art implementations for privacy preserving linear and logistic regressions, and scale to millions of data samples with thousands of features. We also implement the first privacy preserving system for training neural networks.
568 citations
TL;DR: A cloud-based mobileedge computing (MEC) off-loading framework in vehicular networks is proposed, where the tasks are adaptively off-loaded to the MEC servers through direct uploading or predictive relay transmissions, which greatly reduces the cost of computation and improves task transmission efficiency.
Abstract: Cloud-based vehicular networks are a promising paradigm to improve vehicular services through distributing computation tasks between remote clouds and local vehicular terminals. To further reduce the latency and the transmission cost of the computation off-loading, we propose a cloud-based mobileedge computing (MEC) off-loading framework in vehicular networks. In this framework, we study the effectiveness of the computation transfer strategies with vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communication modes. Considering the time consumption of the computation task execution and the mobility of the vehicles, we present an efficient predictive combination-mode relegation scheme, where the tasks are adaptively off-loaded to the MEC servers through direct uploading or predictive relay transmissions. Illustrative results indicate that our proposed scheme greatly reduces the cost of computation and improves task transmission efficiency.
550 citations
14 Oct 2017
TL;DR: An extensive characterization of Microsoft Azure's VM workload, including distributions of the VMs' lifetime, deployment size, and resource consumption is introduced, and Resource Central, a system that collects VM telemetry, learns these behaviors offline, and provides predictions online to various resource managers via a general client-side library is introduced.
Abstract: Cloud research to date has lacked data on the characteristics of the production virtual machine (VM) workloads of large cloud providers. A thorough understanding of these characteristics can inform the providers' resource management systems, e.g. VM scheduler, power manager, server health manager. In this paper, we first introduce an extensive characterization of Microsoft Azure's VM workload, including distributions of the VMs' lifetime, deployment size, and resource consumption. We then show that certain VM behaviors are fairly consistent over multiple lifetimes, i.e. history is an accurate predictor of future behavior. Based on this observation, we next introduce Resource Central (RC), a system that collects VM telemetry, learns these behaviors offline, and provides predictions online to various resource managers via a general client-side library. As an example of RC's online use, we modify Azure's VM scheduler to leverage predictions in oversubscribing servers (with oversubscribable VM types), while retaining high VM performance. Using real VM traces, we then show that the prediction-informed schedules increase utilization and prevent physical resource exhaustion. We conclude that providers can exploit their workloads' characteristics and machine learning to improve resource management substantially.
479 citations
14 Oct 2017
TL;DR: This work presents NetCache, a new key-value store architecture that leverages the power and flexibility of new-generation programmable switches to handle queries on hot items and balance the load across storage nodes, and shows that it improves the throughput by 3-10x and reduces the latency of up to 40% of queries by 50%, for high-performance, in-memory key- value stores.
Abstract: We present NetCache, a new key-value store architecture that leverages the power and flexibility of new-generation programmable switches to handle queries on hot items and balance the load across storage nodes. NetCache provides high aggregate throughput and low latency even under highly-skewed and rapidly-changing workloads. The core of NetCache is a packet-processing pipeline that exploits the capabilities of modern programmable switch ASICs to efficiently detect, index, cache and serve hot key-value items in the switch data plane. Additionally, our solution guarantees cache coherence with minimal overhead. We implement a NetCache prototype on Barefoot Tofino switches and commodity servers and demonstrate that a single switch can process 2+ billion queries per second for 64K items with 16-byte keys and 128-byte values, while only consuming a small portion of its hardware resources. To the best of our knowledge, this is the first time that a sophisticated application-level functionality, such as in-network caching, has been shown to run at line rate on programmable switches. Furthermore, we show that NetCache improves the throughput by 3-10x and reduces the latency of up to 40% of queries by 50%, for high-performance, in-memory key-value stores.
437 citations
21 Jul 2017
TL;DR: In this article, the authors proposed a mechanism that combines data deduplication with dynamic data operations in the privacy preserving public auditing for secure cloud storage, which is highly efficient and provably secure.
Abstract: Cloud storage service has been increasing in popularity as cloud computing plays an important role in the IT domain. Users can be relieved of the burden of storage and computation, by outsourcing the large data files to the cloud servers. However, from the cloud service providers' point of view, it is wise to utilize the data deduplication techniques to reduce the costs of running large storage system and energy consumption on cloud servers. Based on the dynamic nature of data in the cloud storage system, we not only need to assure the data integrity with an auditing protocol supporting dynamic data operations for users, but also consider resorting to data deduplication techniques in the dynamic data operations for cloud service providers to achieve the goal of reducing costs. Thus, in this paper, we propose a mechanism that combines data deduplication with dynamic data operations in the privacy preserving public auditing for secure cloud storage. The analysis of security and performance shows that the proposed mechanism is highly efficient and provably secure.
07 Aug 2017
TL;DR: The system, called SilkRoad, is defined in a 400 line P4 program and when compiled to a state-of-the-art switching ASIC, it can load-balance ten million connections simultaneously at line rate.
Abstract: In this paper, we show that up to hundreds of software load balancer (SLB) servers can be replaced by a single modern switching ASIC, potentially reducing the cost of load balancing by over two orders of magnitude. Today, large data centers typically employ hundreds or thousands of servers to load-balance incoming traffic over application servers. These software load balancers (SLBs) map packets destined to a service (with a virtual IP address, or VIP), to a pool of servers tasked with providing the service (with multiple direct IP addresses, or DIPs). An SLB is stateful, it must always map a connection to the same server, even if the pool of servers changes and/or if the load is spread differently across the pool. This property is called per-connection consistency or PCC. The challenge is that the load balancer must keep track of millions of connections simultaneously.Until recently, it was not possible to implement a load balancer with PCC in a merchant switching ASIC, because high-performance switching ASICs typically can not maintain per-connection states with PCC. Newer switching ASICs provide resources and primitives to enable PCC at a large scale. In this paper, we explore how to use switching ASICs to build much faster load balancers than have been built before. Our system, called SilkRoad, is defined in a 400 line P4 program and when compiled to a state-of-the-art switching ASIC, we show it can load-balance ten million connections simultaneously at line rate.
TL;DR: This paper presents a method for minimizing Service Delay in a scenario with two cloudlet servers, which has a dual focus on computation and communication elements, controlling Processing Delay through virtual machine migration and improving Transmission Delay with Transmission Power Control.
Abstract: Due to physical limitations, mobile devices are restricted in memory, battery, processing, among other characteristics. This results in many applications that cannot be run in such devices. This problem is fixed by Edge Cloud Computing, where the users offload tasks they cannot run to cloudlet servers in the edge of the network. The main requirement of such a system is having a low Service Delay, which would correspond to a high Quality of Service. This paper presents a method for minimizing Service Delay in a scenario with two cloudlet servers. The method has a dual focus on computation and communication elements, controlling Processing Delay through virtual machine migration and improving Transmission Delay with Transmission Power Control. The foundation of the proposal is a mathematical model of the scenario, whose analysis is used on a comparison between the proposed approach and two other conventional methods; these methods have single focus and only make an effort to improve either Transmission Delay or Processing Delay, but not both. As expected, the proposal presents the lowest Service Delay in all study cases, corroborating our conclusion that a dual focus approach is the best way to tackle the Service Delay problem in Edge Cloud Computing.
TL;DR: An integrated framework for computation offloading and interference management in wireless cellular networks with MEC is proposed and the outcomes of the offloading decision and PRB allocation are used to distribute the computation resource of the MEC server to the UEs.
Abstract: Mobile edge computing (MEC) has attracted great interests as a promising approach to augment computational capabilities of mobile devices. An important issue in the MEC paradigm is computation offloading. In this paper, we propose an integrated framework for computation offloading and interference management in wireless cellular networks with MEC. In this integrated framework, we formulate the computation offloading decision, physical resource block (PRB) allocation, and MEC computation resource allocation as optimization problems. The MEC server makes the offloading decision according to the local computation overhead estimated by all user equipments (UEs) and the offloading overhead estimated by the MEC server itself. Then, the MEC server performs the PRB allocation using the graph coloring method. The outcomes of the offloading decision and PRB allocation are then used to distribute the computation resource of the MEC server to the UEs. Simulation results are presented to show the effectiveness of the proposed scheme with different system parameters.
TL;DR: The proposed consolidation algorithm is based on a migration policy of VNFIs that considers the revenue loss due to QoS degradation that a user suffers due to information loss occurring during the migrations.
Abstract: Network function virtualization foresees the virtualization of service functions and their execution on virtual machines. Any service is represented by a service function chain (SFC) that is a set of VNFs to be executed according to a given order. The running of VNFs needs the instantiation of VNF Instances (VNFIs) that in general are software modules executed on virtual machines. The virtualization challenges include: 1) where to instantiate VNFIs; ii) how many resources to allocate to each VNFI; iii) how to route SFC requests to the appropriate VNFIs in the right sequence; and iv) when and how to migrate VNFIs in response to changes to SFC request intensity and location. We develop an approach that uses three algorithms that are used back-to-back resulting in VNFI placement, SFC routing, and VNFI migration in response to changing workload. The objective is to first minimize the rejection of SFC bandwidth and second to consolidate VNFIs in as few servers as possible so as to reduce the energy consumed. The proposed consolidation algorithm is based on a migration policy of VNFIs that considers the revenue loss due to QoS degradation that a user suffers due to information loss occurring during the migrations. The objective is to minimize the total cost given by the energy consumption and the revenue loss due to QoS degradation. We evaluate our suite of algorithms on a test network and show performance gains that can be achieved over using other alternative naive algorithms.
Posted Content•
TL;DR: In this paper, the problem of joint task offloading and resource allocation in a MEC enabled multi-cell wireless network is considered, where each base station (BS) is equipped with a mobile edge computing server that can assist mobile users in executing computation-intensive tasks via offloading.
Abstract: Mobile-Edge Computing (MEC) is an emerging paradigm that provides a capillary distribution of cloud computing capabilities to the edge of the wireless access network, enabling rich services and applications in close proximity to the end users. In this article, a MEC enabled multi-cell wireless network is considered where each Base Station (BS) is equipped with a MEC server that can assist mobile users in executing computation-intensive tasks via task offloading. The problem of Joint Task Offloading and Resource Allocation (JTORA) is studied in order to maximize the users' task offloading gains, which is measured by the reduction in task completion time and energy consumption. The considered problem is formulated as a Mixed Integer Non-linear Program (MINLP) that involves jointly optimizing the task offloading decision, uplink transmission power of mobile users, and computing resource allocation at the MEC servers. Due to the NP-hardness of this problem, solving for optimal solution is difficult and impractical for a large-scale network. To overcome this drawback, our approach is to decompose the original problem into (i) a Resource Allocation (RA) problem with fixed task offloading decision and (ii) a Task Offloading (TO) problem that optimizes the optimal-value function corresponding to the RA problem. We address the RA problem using convex and quasi-convex optimization techniques, and propose a novel heuristic algorithm to the TO problem that achieves a suboptimal solution in polynomial time. Numerical simulation results show that our algorithm performs closely to the optimal solution and that it significantly improves the users' offloading utility over traditional approaches.
01 May 2017
TL;DR: The first online job dispatching and scheduling algorithm in edge-clouds, called OnDisc, is derived, which is scalable in the speed augmentation model; that is, OnDisc is (1 + ε)-speed O(1/ε)-competitive for any constant ε ∊ (0,1).
Abstract: 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 + e)-speed O(1/e)-competitive for any constant e ∊ (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.
21 May 2017
TL;DR: This work proposes a hierarchical cloud-based Vehicular Edge Computing (VEC) offloading framework, where a backup computing server in the neighborhood is introduced to make up for the deficit computing resources of MEC servers.
Abstract: The increasing number of smart vehicles and their resource hungry applications pose new challenges in terms of computation and processing for providing reliable and efficient vehicular services. Mobile Edge Computing (MEC) is a new paradigm with potential to improve vehicular services through computation offloading in close proximity to mobile vehicles. However, in the road with dense traffic flow, the computation limitation of these MEC servers may endanger the quality of offloading service. To address the problem, we propose a hierarchical cloud-based Vehicular Edge Computing (VEC) offloading framework, where a backup computing server in the neighborhood is introduced to make up for the deficit computing resources of MEC servers. Based on this framework, we adopt a Stackelberg game theoretic approach to design an optimal multilevel offloading scheme, which maximizes the utilities of both the vehicles and the computing servers. Furthermore, to obtain the optimal offloading strategies, we present an iterative distributed algorithm and prove its convergence. Numerical results indicate that our proposed scheme greatly enhances the utility of the offloading service providers.
TL;DR: A system that includes NB devices, an IoT cloud platform, an application server, and a user app is designed that provides an easy approach to academic research as well as commercial applications.
Abstract: Recently, narrowband Internet of Things (NB-IoT), one of the most promising low power wide area (LPWA) technologies, has attracted much attention from both academia and industry. It has great potential to meet the huge demand for machine-type communications in the era of IoT. To facilitate research on and application of NB-IoT, in this paper, we design a system that includes NB devices, an IoT cloud platform, an application server, and a user app. The core component of the system is to build a development board that integrates an NB-IoT communication module and a subscriber identification module, a micro-controller unit and power management modules. We also provide a firmware design for NB device wake-up, data sensing, computing and communication, and the IoT cloud configuration for data storage and analysis. We further introduce a framework on how to apply the proposed system to specific applications. The proposed system provides an easy approach to academic research as well as commercial applications.
TL;DR: An online algorithm to learn the unknown dynamic environment and guarantee that the performance gap compared to the optimal strategy is bounded by a logarithmic function with time is proposed.
Abstract: With mobile devices increasingly able to connect to cloud servers from anywhere, resource-constrained devices can potentially perform offloading of computational tasks to either save local resource usage or improve performance. It is of interest to find optimal assignments of tasks to local and remote devices that can take into account the application-specific profile, availability of computational resources, and link connectivity, and find a balance between energy consumption costs of mobile devices and latency for delay-sensitive applications. We formulate an NP-hard problem to minimize the application latency while meeting prescribed resource utilization constraints. Different from most of existing works that either rely on the integer programming solver, or on heuristics that offer no theoretical performance guarantees, we propose Hermes, a novel fully polynomial time approximation scheme (FPTAS). We identify for a subset of problem instances, where the application task graphs can be described as serial trees, Hermes provides a solution with latency no more than $(1+\epsilon)$ times of the minimum while incurring complexity that is polynomial in problem size and $\frac{1}{\epsilon}$ . We further propose an online algorithm to learn the unknown dynamic environment and guarantee that the performance gap compared to the optimal strategy is bounded by a logarithmic function with time. Evaluation is done by using real data set collected from several benchmarks, and is shown that Hermes improves the latency by $16$ percent compared to a previously published heuristic and increases CPU computing time by only $0.4$ percent of overall latency.
05 Jun 2017
TL;DR: The emerging deep reinforcement learning (DRL) technique, which can deal with complicated control problems with large state space, is adopted to solve the global tier problem and the proposed framework can achieve the best trade-off between latency and power/energy consumption in a server cluster.
Abstract: Automatic decision-making approaches, such as reinforcement learning (RL), have been applied to (partially) solve the resource allocation problem adaptively in the cloudcomputing system. However, a complete cloud resource allocation framework exhibits high dimensions in state and action spaces, which prohibit the usefulness of traditional RL techniques. In addition, high power consumption has become one of the critical concerns in design and control of cloud computing systems, which degrades system reliability and increases cooling cost. An effective dynamic power management (DPM) policy should minimize power consumption while maintaining performance degradationwithin an acceptable level. Thus, a joint virtual machine (VM) resource allocation and power management framework is critical to the overall cloud computing system. Moreover, novel solution framework is necessary to address the even higher dimensions in state and action spaces. In this paper, we propose a novel hierarchical framework forsolving the overall resource allocation and power management problem in cloud computing systems. The proposed hierarchical framework comprises a global tier for VM resource allocation to the servers and a local tier for distributed power management of local servers. The emerging deep reinforcement learning (DRL) technique, which can deal with complicated control problems with large state space, is adopted to solve the global tier problem. Furthermore, an autoencoder and a novel weight sharing structure are adopted to handle the high-dimensional state space and accelerate the convergence speed. On the other hand, the local tier of distributed server power managements comprises an LSTM based workload predictor and a model-free RL based power manager, operating in a distributed manner. Experiment results using actual Google cluster traces showthat our proposed hierarchical framework significantly savesthe power consumption and energy usage than the baselinewhile achieving no severe latency degradation. Meanwhile, the proposed framework can achieve the best trade-off between latency and power/energy consumption in a server cluster.
TL;DR: A fundamentally different approach is needed, in which the cache contents are used as side information for coded communication over the shared link, and it is proposed and proved that it is close to optimal.
Abstract: We consider a network consisting of a file server connected through a shared link to a number of users, each equipped with a cache. Knowing the popularity distribution of the files, the goal is to optimally populate the caches, such as to minimize the expected load of the shared link. For a single cache, it is well known that storing the most popular files is optimal in this setting. However, we show here that this is no longer the case for multiple caches. Indeed, caching only the most popular files can be highly suboptimal. Instead, a fundamentally different approach is needed, in which the cache contents are used as side information for coded communication over the shared link. We propose such a coded caching scheme and prove that it is close to optimal.
25 Jun 2017
TL;DR: A general framework for IoT-fog-cloud applications is introduced, and a delay-minimizing policy for fog-capable devices that aims to reduce the service delay for IoT applications is proposed, and how the proposed framework helps to reduce IoT service delay is shown.
Abstract: With the Internet of Things (IoT) becoming a major component of our daily life, understanding how to improve quality of service (QoS) in IoT networks is becoming a challenging problem. Currently most interaction between the IoT devices and the supporting back-end servers is done through large scale cloud data centers. However, with the exponential growth of IoT devices and the amount of data they produce, communication between "things" and cloud will be costly, inefficient, and in some cases infeasible. Fog computing serves as solution for this as it provides computation, storage, and networking resource for IoT, closer to things and users. One of the promising advantages of fog is reducing service delay for end user applications, whereas cloud provides extensive computation and storage capacity with a higher latency. Thus it is necessary to understand the interplay between fog computing and cloud, and to evaluate the effect of fog computing on the IoT service delay and QoS. In this paper we will introduce a general framework for IoT-fog-cloud applications, and propose a delay-minimizing policy for fog-capable devices that aims to reduce the service delay for IoT applications. We then develop an analytical model to evaluate our policy and show how the proposed framework helps to reduce IoT service delay.
TL;DR: A regional cooperative fog-computing-based intelligent vehicular network (CFC-IoV) architecture for dealing with big IoV data in the smart city is proposed, including mobility control, multi-source data acquisition, distributed computation and storage, and multi-path data transmission.
Abstract: As vehicle applications, mobile devices and the Internet of Things are growing fast, and developing an efficient architecture to deal with the big data in the Internet of Vehicles (IoV) has been an important concern for the future smart city. To overcome the inherent defect of centralized data processing in cloud computing, fog computing has been proposed by offloading computation tasks to local fog servers (LFSs). By considering factors like latency, mobility, localization, and scalability, this article proposes a regional cooperative fog-computing-based intelligent vehicular network (CFC-IoV) architecture for dealing with big IoV data in the smart city. Possible services for IoV applications are discussed, including mobility control, multi-source data acquisition, distributed computation and storage, and multi-path data transmission. A hierarchical model with intra-fog and inter-fog resource management is presented, and energy efficiency and packet dropping rates of LFSs in CFC-IoV are optimized.
TL;DR: This work compares two VM mobility modes, bulk and live migration, as a function of mobile cloud service requirements, determining that a high preference should be given to live migration and bulk migrations seem to be a feasible alternative on delay-stringent tiny-disk services, such as augmented reality support, and only with further relaxation on network constraints.
Abstract: Major interest is currently given to the integration of clusters of virtualization servers, also referred to as ‘cloudlets’ or ‘edge clouds’, into the access network to allow higher performance and reliability in the access to mobile edge computing services. We tackle the edge cloud network design problem for mobile access networks. The model is such that the virtual machines (VMs) are associated with mobile users and are allocated to cloudlets. Designing an edge cloud network implies first determining where to install cloudlet facilities among the available sites, then assigning sets of access points, such as base stations to cloudlets, while supporting VM orchestration and considering partial user mobility information, as well as the satisfaction of service-level agreements. We present link-path formulations supported by heuristics to compute solutions in reasonable time. We qualify the advantage in considering mobility for both users and VMs as up to 20% less users not satisfied in their SLA with a little increase of opened facilities. We compare two VM mobility modes, bulk and live migration, as a function of mobile cloud service requirements, determining that a high preference should be given to live migration, while bulk migrations seem to be a feasible alternative on delay-stringent tiny-disk services, such as augmented reality support, and only with further relaxation on network constraints.
TL;DR: In this paper, a general framework for PIR from arbitrary coded databases is presented, which allows one to adjust the rate of the scheme to the suspected number of colluding servers.
Abstract: We present a general framework for private information retrieval (PIR) from arbitrary coded databases that allows one to adjust the rate of the scheme to the suspected number of colluding servers. ...
TL;DR: A distributed reputation management system (DREAMS) is proposed, wherein VEC servers are adopted to execute local reputation management tasks for vehicles, and the effectiveness of the reputation-based resource allocation algorithm is demonstrated.
Abstract: Vehicular edge computing (VEC) is introduced to extend computing capacity to vehicular network edge recently. With the advent of VEC, service providers directly host services in close proximity of mobile vehicles for great improvements. As a result, a new networking paradigm, vehicular edge networks is emerged along with the development of VEC. However, it is necessary to address security issues for facilitating VEC well. In this paper, we focus on reputation management to ensure security protection and improve network efficiency in the implementation of VEC. A distributed reputation management system ( DREAMS ) is proposed, wherein VEC servers are adopted to execute local reputation management tasks for vehicles. This system has remarkable features for improving overall performance: 1) distributed reputation maintenance; 2) trusted reputation manifestation; 3) accurate reputation update; and 4) available reputation usage. In particular, we utilize multi-weighted subjective logic for accurate reputation update in DREAMS. To enrich reputation usage in DREAMS, service providers optimize resource allocation in computation offloading by considering reputation of vehicles. Numerical results indicate that DREAMS has great advantages in optimizing misbehavior detection and improving the recognition rate of misbehaving vehicles. Meanwhile, we demonstrate the effectiveness of our reputation-based resource allocation algorithm.
09 May 2017
TL;DR: A heterogeneity-aware algorithm that uses a constant learning rate schedule for updates before adding them to the global parameter allows us to suppress stragglers' harm on robust convergence and theoretically prove the valid convergence of both approaches.
Abstract: We study distributed machine learning in heterogeneous environments in this work. We first conduct a systematic study of existing systems running distributed stochastic gradient descent; we find that, although these systems work well in homogeneous environments, they can suffer performance degradation, sometimes up to 10x, in heterogeneous environments where stragglers are common because their synchronization protocols cannot fit a heterogeneous setting. Our first contribution is a heterogeneity-aware algorithm that uses a constant learning rate schedule for updates before adding them to the global parameter. This allows us to suppress stragglers' harm on robust convergence. As a further improvement, our second contribution is a more sophisticated learning rate schedule that takes into consideration the delayed information of each update. We theoretically prove the valid convergence of both approaches and implement a prototype system in the production cluster of our industrial partner Tencent Inc. We validate the performance of this prototype using a range of machine-learning workloads. Our prototype is 2-12x faster than other state-of-the-art systems, such as Spark, Petuum, and TensorFlow; and our proposed algorithm takes up to 6x fewer iterations to converge.
01 May 2017
TL;DR: An efficient three-step algorithm comprising of semidefinite relaxation (SDR), alternating optimization (AO), and sequential tuning (ST) is shown to always compute a locally optimal solution, and give nearly optimal performance under a wide range of parameter settings.
Abstract: We consider a general multi-user mobile cloud computing system with a computing access point (CAP), where each mobile user has multiple independent tasks that may be processed locally, at the CAP, or at a remote cloud server. The CAP serves both as the network access gateway and a computation service provider to the mobile users. We aim to jointly optimize the offloading decisions of all users' tasks as well as the allocation of computation and communication resources, to minimize the overall cost of energy, computation, and delay for all users. This problem is NP-hard in general. We propose an efficient three-step algorithm comprising of semidefinite relaxation (SDR), alternating optimization (AO), and sequential tuning (ST). It is shown to always compute a locally optimal solution, and give nearly optimal performance under a wide range of parameter settings. Through evaluating the performance of different combinations of the three components of this SDR-AO-ST algorithm, we provide insights into their roles and contributions in the overall solution. We further compare the performance of SDR-AO-ST against a lower bound to the minimum cost, purely local processing, purely cloud processing, and hybrid local-cloud processing without using the CAP. Our numerical results demonstrate the effectiveness of the proposed algorithm in the joint management of computation and communication resources in mobile cloud computing systems with a CAP.
19 Mar 2017
TL;DR: Simulation results show that task offloading scheduling is more critical when the available radio and computational resources in MEC systems are relatively balanced, and it is shown that the proposed algorithm achieves near-optimal execution delay along with a substantial device energy saving.
Abstract: Mobile-edge computing (MEC) has emerged as a prominent technique to provide mobile services with high computation requirement, by migrating the computation- intensive tasks from the mobile devices to the nearby MEC servers. To reduce the execution latency and device energy consumption, in this paper, we jointly optimize task offloading scheduling and transmit power allocation for MEC systems with multiple independent tasks. A low-complexity sub-optimal algorithm is proposed to minimize the weighted sum of the execution delay and device energy consumption based on alternating minimization. Specifically, given the transmit power allocation, the optimal task offloading scheduling, i.e., to determine the order of offloading, is obtained with the help of flow shop scheduling theory. Besides, the optimal transmit power allocation with a given task offloading scheduling decision will be determined using convex optimization techniques. Simulation results show that task offloading scheduling is more critical when the available radio and computational resources in MEC systems are relatively balanced. In addition, it is shown that the proposed algorithm achieves near-optimal execution delay along with a substantial device energy saving.