With advances in wireless communication technology, more and more people depend heavily on portable mobile devices for business, entertainments and social interactions. This poses a great challenge of building a seamless application experience across different computing platforms. A key issue is the resource limitations of mobile devices due to their portable size, however this can be overcome by offloading computation-intensive tasks from the mobile devices to clusters of nearby computers called cloudlets through wireless access points. As increasing numbers of people access the Internet via mobile devices, it is reasonable to envision in the near future that cloudlet services will be available for the public through easily accessible public wireless metropolitan area networks (WMANs). However, the outdated notion of treating cloudlets as isolated data-centers-in-boxes must be discarded as there are clear benefits to connecting multiple cloudlets together to form a network. In this paper we investigate how to balance the workload among cloudlets in an WMAN to optimize mobile application performance. We first introduce a novel system model to capture the response time delays of offloaded tasks and formulate an optimization problem with the aim to minimize the maximum response time of all offloaded tasks. We then propose two algorithms for the problem: one is a fast heuristic, and another is a distributed genetic algorithm that is capable of delivering a more accurate solution compared with the first algorithm, but at the expense of a much longer running time. We finally evaluate the performance of the proposed algorithms in realistic simulation environments. The experimental results demonstrate the significant potential of the proposed algorithms in reducing the user task response time, maximizing user experience.

Multicasting is a fundamental functionality of many network applications, including online conferencing, event monitoring, video streaming, and so on. To ensure reliable, secure, and scalable multicasting, a service chain that consists of network functions (e.g., firewalls, intrusion detection systems, and transcoders) usually is associated with each multicast request. We refer to such a multicast request with service chain requirement as an network function virtualization (NFV)-enabled multicast request. In this paper, we study NFV-enabled multicasting in a software-defined network (SDN) with an aim to maximize network throughput while minimizing the implementation cost of admitted NFV-enabled multicast requests, subject to network resource capacity, where the implementation cost of a request consists of its computing resource consumption cost in servers and its network bandwidth consumption cost when routing and processing its data packets in the network. To this end, we first formulate two NFV-enabled multicasting problems with and without resource capacity constraints and one online NFV-enabled multicasting problem. We then devise two approximation algorithms with an approximation ratio of $2M$ for the NFV-enabled multicasting problems with and without resource capacity constraints, if the number of servers for implementing the service chain of each request is no greater than a constant $M$ (≥1). We also study dynamic admissions of NFV-enabled multicast requests without the knowledge of future request arrivals with the objective to maximize the network throughput, for which we propose an efficient heuristic, and for the special case of dynamic request admissions, we devise an online algorithm with a competitive ratio of $O(\log n)$ for it when $M=1$ , where $n$ is the number of nodes in the network. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results demonstrate that the proposed algorithms are promising and outperform existing heuristics.

Pushing the cloud frontier to the network edge close to mobile users has attracted tremendous interest not only from cloud operators but also from network service providers. In particular, the deployment of cloudlets in metropolitan area networks enables network service providers to provide low-latency services to mobile users through implementing their specified virtualized network functions (VNFs) while meeting their Quality-of-Service (QoS) requirements. In this paper, we formulate a novel task offloading problem in a mobile edge-cloud network, where each offloading task requests a specified network function with a tolerable delay. We aim to maximize the number of requests admitted while minimizing the operational cost of admitted requests within a finite time horizon, through either sharing existing VNF instances or creating new VNF instances in cloudlets. We first show that the problem is NP-hard, and then devise an efficient online algorithm for the problem by reducing it to a series of minimum weight maximum matching problems. Considering dynamic changes of task offloading request patterns over time, we further develop an effective prediction mechanism for new VNF instance creations and idle VNF instance releases to further lower the operational cost of the network service provider. Also, we devise an online algorithm with a competitive ratio for a special case of the problem where the delay requirements of requests are negligible. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results indicate that the proposed algorithms are promising.

Mobile Edge Computing (MEC) is essential for enabling new innovative technologies that depend on low-latency computation environments such as Augmented Reality (AR). As AR applications continue to deliver better graphics with richer interactive features, AR devices will increasingly rely on nearby cloudlets to assist with the demanding computation requirements of AR applications. Supporting multiplayer interactions in an MEC environment brings many challenges. Processing user interactions can be computation-intensive especially when multiple users in close proximity to each other are acting simultaneously; the limited resources of a cloudlet could be overwhelmed if there are too many players involved. In this paper, we envision a scenario in the near future where players wearing AR heads-up display devices engage with other players over a large area with densely deployed cloudlets. We first propose a novel system model, and then formulate the Decentralized Multiplayer Coordination (DMC) Problem with the aim of minimizing the game frame duration among players, and devise an efficient algorithm for the problem. We finally evaluate the performance of the proposed algorithm through experimental simulations. Experimental results demonstrate that the proposed algorithm is promising.

Data transfer in contemporary networks usually is associated with strict policy enforcement for data transfer security and system performance purposes. Such a policy is represented by a service chain consisting of a sequence of network functions such as firewalls, intrusion detection systems, transcoders, etc. Due to the high cost and inflexibility of managing hardware-based network functions, network function virtualization (NFV) has emerged as a promising technology to meet the stringent requirement imposed on the service chain of each data transfer request in a low-cost and flexible way. In this paper, we study policy-aware unicast request admissions with and without end-to-end delay constraints in a software defined network. We aim to minimize the operational cost of admitting a single request in terms of both computing resource consumption for implementing the NFVs in the service chain and bandwidth resource consumption for routing its data traffic, or maximize the network throughput for a sequence of requests without the knowledge of future request arrivals. We first formulate four novel optimization problems and provide a generic optimization framework for the problems. We then develop efficient algorithms for the admission of a single NFV-enabled request with and without the end-to-end delay constraint, where NFV-enabled requests are defined as the requests with policy enforcement requirements. We also devise online algorithms with a guaranteed performance for dynamic admissions of requests without the knowledge of future arrivals. In particular, we provide the very first online algorithm with a provable competitive ratio for the problem without the end-to-end delay requirement. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results show that the proposed algorithms are promising and outperform existing heuristics.

Data transfers in contemporary networks depend upon network functions for ensuring data security and system performance. These policies are represented by a service chain that consists of different network functions such as firewalls, Intrusion Detection Systems (IDSs), transcoders, etc. Network Function Virtualization (NFV) has emerged as a promising technology to meet the stringent requirement imposed on the service chain. In this paper, we study NFV-enabled unicasting in SDNs with and without end-to-end delay constraints. We aim to maximize the network throughput for a sequence of NFV-enabled unicast requests without the knowledge of future arrivals. We first formulate the problems as novel optimization problems in terms of both computing and bandwidth resource consumptions, and provide a generic optimization framework. We then develop an online algorithm with guaranteed performance without the delay requirement and a heuristic with the delay requirement. We finally evaluate the performance of the proposed algorithms through experimental simulations. The results of the experimental simulations show that the proposed algorithms are promising.

Multicasting is a fundamental functionality of networks for many applications including online conferencing, event monitoring, video streaming, and system monitoring in data centers. To ensure multicasting reliable, secure and scalable, a service chain consisting of network functions (e.g., firewalls, Intrusion Detection Systems (IDSs), and transcoders) usually is associated with each multicast request. Such a multicast request is referred to as an NFV-enabled multicast request. In this paper we study NFV-enabled multicasting in a Software-Defined Network (SDN) with the aims to minimize the implementation cost of each NFV-enabled multicast request or maximize the network throughput for a sequence of NFV-enabled requests, subject to network resource capacity constraints. We first formulate novel NFV-enabled multicasting and online NFV-enabled multicasting problems. We then devise the very first approximation algorithm with an approximation ratio of 2K for the NFV-enabled multicasting problem if the number of servers for implementing the network functions of each request is no more than a constant K (1). We also study dynamic admissions of NFV-enabled multicast requests without the knowledge of future request arrivals with the objective to maximize the network throughput, for which we propose an online algorithm with a competitive ratio of O(log n) when K = 1, where n is the number of nodes in the network. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results demonstrate that the proposed algorithms outperform other existing heuristics.

Pushing the cloud frontier to the network edge has attracted tremendous interest not only from cloud operators of the IT service/software industry but also from network service operators that provide various network services for mobile users. In particular, by deploying cloudlets in metropolitan area networks, network service providers can provide various network services through implementing virtualized network functions to meet the demands of mobile users. In this paper we formulate a novel task offloading problem in a metropolitan area network, where each offloaded task requests a specific network function with a maximum tolerable delay and different offloading requests may require different network services. We aim to maximize the number of requests admitted while minimizing their admission cost within a finite time horizon. We first show that the problem is NP-hard, and then devise an efficient algorithm through reducing the problem to a series of minimum eight maximum matching in auxiliary bipartite graphs. We also consider dynamic changes of offloading request patterns over time, and develop an effective prediction mechanism to release and/or create instances of network functions in different cloudlets for cost savings. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results indicate that the proposed algorithms are promising.

Mobile applications are becoming increasingly computation-intensive, while the computing capability of portable mobile devices is limited. A powerful way to reduce the completion time of an application in a mobile device is to offload its tasks to nearby cloudlets, which consist of clusters of computers. Although there is a significant body of research in mobile cloudlet offloading technology, there has been very little attention paid to how cloudlets should be placed in a given network to optimize mobile application performance. In this paper we study cloudlet placement and mobile user allocation to the cloudlets in a wireless metropolitan area network (WMAN). We devise an algorithm for the problem, which enables the placement of the cloudlets at user dense regions of the WMAN, and assigns mobile users to the placed cloudlets while balancing their workload. We also conduct experiments through simulation. The simulation results indicate that the performance of the proposed algorithm is very promising.

With advances in wireless communication technology, more and more people depend heavily on portable mobile devices for businesses, entertainments and social interactions. Although such portable mobile devices can offer various promising applications, their computing resources remain limited due to their portable size. This however can be overcome by remotely executing computation-intensive tasks on clusters of near by computers known as cloudlets. As increasing numbers of people access the Internet via mobile devices, it is reasonable to envision in the near future that cloudlet services will be available for the public through easily accessible public wireless metropolitan area networks (WMANs). However, the outdated notion of treating cloudlets as isolated data-centers-in-a-box must be discarded as there are clear benefits to connecting multiple cloudlets together to form a network. In this paper we investigate how to balance the workload between multiple cloudlets in a network to optimize mobile application performance. We first introduce a system model to capture the response times of offloaded tasks, and formulate a novel optimization problem, that is to find an optimal redirection of tasks between cloudlets such that the maximum of the average response times of tasks at cloudlets is minimized. We then propose a fast, scalable algorithm for the problem. We finally evaluate the performance of the proposed algorithm through experimental simulations. The experimental results demonstrate the significant potential of the proposed algorithm in reducing the response times of tasks.

Today's computer networks rely on a wide spectrum of specialized middleboxes to improve their security and performance. Traditional middleboxes that are implemented by dedicated hardware are expensive and hard to manage. A promising technique of consolidated middleboxes - implementing traditional middleboxes in Virtual Machines (VMs) - offers economical yet simplified management of middleboxes in Software-Defined Networks (SDNs). However there are still challenges to realizing user routing requests with network function enforcement (a sequence of middleboxes) while maximizing the network throughput, due to various resource constraints on SDNs, such as forwarding table capacity at each switch, bandwidth resource capacity at each link, and computing resource capacity at each server (Physical Machine). In this paper, we study the problem of maximizing the network throughput of an SDN by admitting as many user requests as possible, where each user request has both bandwidth and computing resource demands to implement its network functions (consolidated middleboxes). We first formulate the problem as a novel network throughput maximization problem. We then provide an Integer Linear Program (ILP) solution for it if the problem size is small, otherwise, we devise two heuristics that strive for the fine tradeoff between the accuracy of solutions and the running times of achieving the solutions. We finally evaluate the performance of the proposed algorithms by simulations, based on real and synthetic network topologies. Experimental results demonstrate that the proposed algorithms are very promising.

Mobile cloud computing is emerging as a main ubiquitous computing platform to provide rich cloud resources for various applications of mobile devices. Although most existing studies in mobile cloud computing focus on energy savings of mobile devices by offloading computing-intensive jobs from mobile devices to remote clouds, the access delays between mobile users and remote clouds usually are long and sometimes unbearable. Cloudlet as a new technology is capable to bridge this gap, and can enhance the performance of mobile devices significantly while meeting the crisp response time requirements of mobile users. In this paper, we study the cloudlet placement problem in a large-scale Wireless Metropolitan Area Network (WMAN) consisting of many wireless Access Points (APs). We first formulate the problem as a novel capacitated cloudlet placement problem that places$K$cloudlets to some strategic locations in the WMAN with the objective to minimize the average access delay between mobile users and the cloudlets serving the users. We then propose an exact solution to the problem by formulating it as an Integer Linear Programming (ILP). Due to the poor scalability of the ILP, we instead propose an efficient heuristic for the problem. For a special case of the problem where all cloudlets have identical computing capacities, we devise novel approximation algorithms with guaranteed approximation ratios. We also devise an online algorithm for dynamically allocating user requests to different cloudlets, if the$K$cloudlets have already been placed. We finally evaluate the performance of the proposed algorithms through experimental simulations. Simulation results demonstrate that the proposed algorithms are promising and scalable.

In this paper we study the cloudlet placement problem in a large-scale Wireless Metropolitan Area Network (WMAN) that consists of many wireless Access Points (APs). Although most existing studies in mobile cloud computing mainly focus on energy savings of mobile devices by offloading computing-intensive jobs from them to remote clouds, the access delay between mobile users and the clouds usually is large and sometimes unbearable. Cloudlet as a new technology is capable to bridge this gap, and has been demonstrated to enhance the performance of mobile devices significantly while meeting the crisp response time requirements of mobile users. In this paper we consider placing multiple cloudlets with different computing capacities at some strategic local locations in a WMAN to reduce the average cloudlet access delay of mobile users at different APs. We first formulate this problem as a novel capacitated cloudlet placement problem that places K cloudlets to some locations in the WMAN with the objective to minimize the average cloudlet access delay between the mobile users and the cloudlets serving their requests. We then propose a fast yet efficient heuristic. For a special case of the problem where all cloudlets have the identical computing capacity, we devise a novel approximation algorithm with a guaranteed approximation ratio. In addition, We also consider allocating user requests to cloudlets by devising an efficient online algorithm for such an assignment. We finally evaluate the performance of the proposed algorithms through experimental simulations. The simulation results demonstrate that the proposed algorithms are promising and scalable.