Mobile Ad Hoc Networks (MANETs) provide rapidly deployable and self-configuring network capacity required in many critical applications, e.g., battlefields, disaster relief wide area sensing. In this paper we study the problem of efficient data delivery sparse MANETs where partitions can last for a significant period. Previous approaches rely on use either long range communication which leads to rapid draining nodes' limited batteries, or existing node mobility results low rates large...

Recent proposals for network virtualization provide a promising way to overcome the Internet ossification. The key idea of is build diversified support variety services and architectures through shared substrate. A major challenge in assigning substrate resources virtual networks (VN) efficiently on-demand. This paper focuses on two versions VN assignment problem: without reconfiguration (VNA-I) with (VNAII). For VNA-I problem, we develop basic scheme as building block all other advanced...

As technology rapidly progresses, more devices will combine both communication and mobility capabilities. With in devices, we envision a new class of proactive networks that are able to adapt themselves, via physical movement, meet the needs applications. To fully realize these opportunities, effective control device interaction between is needed. In this paper, consider message ferrying (MF) scheme which exploits controlled transport data delay-tolerant networks, where end-to-end paths may...

We investigate the design of a reputation system for decentralized unstructured P2P networks like Gnutella. Having reliable information about peers can form basis an incentive and guide in their decision making (e.g., who to download file from). The uses objective criteria track each peer's contribution allows store reputations locally. Reputation are computed using either two schemes, debit-credit computation (DCRC) credit-only (CORC). Using agent (RCA), we public key based mechanism that...

An ad hoc network allows devices with wireless interfaces to communicate each other without any pre-installed infrastructure. Due node mobility, limited radio power, failure and wide deployment area, networks are often vulnerable partitioning. A number of examples in battlefield, disaster recovery area surveillance. Unfortunately, most existing routing protocols will fail deliver messages under these circumstances since no route the destination exists. In this work we propose Message...

Mobile devices are increasingly being relied on for services that go beyond simple connectivity and require more complex processing. Fortunately, a mobile device encounters, possibly intermittently, many entities capable of lending it computational resources. At one extreme is the traditional cloud-computing context where connected to remote cloud resources maintained by service provider with which has an established relationship. In this paper we consider other extreme, device's contacts...

Network function virtualization foresees the of service functions and their execution on virtual machines. Any is represented by a chain (SFC) that set VNFs to be executed according given order. The running needs instantiation VNF Instances (VNFIs) in general are software modules challenges include: 1) where instantiate VNFIs; ii) how many resources allocate each VNFI; iii) route SFC requests appropriate VNFIs right sequence; iv) when migrate response changes request intensity location. We...

In opportunistic networks, end-to-end paths between two communicating nodes are rarely available. such situations, the might still copy and forward messages to that more likely meet destination. The question is which forwarding algorithm offers best trade off cost (number of message replicas) rate successful delivery. We address this challenge by developing PeopleRank approach in ranked using a tunable weighted social information. Similar PageRank idea, gives higher weight if they socially...

Mobile devices are becoming increasingly capable computing platforms with significant processor power and memory. However, mobile compute capabilities often underutilized. In this paper we consider how a collection of co-located can be orchestrated to provide cloud service at the edge. Scenarios include, but not limited to, passengers using public transit services, students in classrooms groups people sitting coffee shop. To end, propose femtocloud system which provides dynamic,...

In typical proposals for video-on-demand (VoD) systems, customers are serviced individually by allocating and dedicating a transmission channel set of server resources to each customer. This approach leads an expensive operate, nonscalable system. We consider VoD system that uses multicast delivery service multiple with single resources. The use communication requires part the on-demand nature be sacrificed achieve scalability cost-effectiveness. One drawback using is it complicates...

Message ferrying is a networking paradigm where special node, called message ferry, facilitates the connectivity in mobile ad hoc network nodes are sparsely deployed. One of key challenges under this design ferry routes to achieve certain properties end-to-end connectivity, such as, delay and loss among network. This difficult problem when move arbitrarily. As we cannot be location nodes, route can contact with certainty. Due difficulty, prior work has either considered for networks...

Disruption tolerant networks (DTNs) are designed to overcome limitations in connectivity due conditions such as mobility, poor infrastructure, and short range radios. DTNs rely on the inherent mobility network deliver packets around frequent extended partitions using a store-carry-and-forward paradigm. However, missed contact opportunities decrease throughput increase delay network. We propose use of throwboxes mobile create greater number opportunities, consequently improving performance...

The routing infrastructure of the Internet has become resistant to fundamental changes and use overlay networks been proposed provide additional flexibility control. One most prominent configurable components an network is its topology, which can be dynamically reconfigured accommodate communication requirements that vary over time. In this paper, we study problem determining dynamic topology reconfiguration for service with requirement, ideal goal find optimal policies minimize potential...

Server replication improves the ability of a service to handle large number clients. One important factors in efficient utilization replicated servers is direct client requests "best" server, according some optimality criteria. In anycasting communication paradigm, sender communicates with receiver chosen from an anycast group equivalent receivers. As such, well suited problem directing clients servers. This paper examines definition and support paradigm at application-layer, providing that...

Delay tolerant networks (DTNs) are a class of emerging that experience frequent and long-duration partitions. These have variety applications in situations such as crisis environments deep-space communication. In this paper, we study the problem multicasting DTNs. Multicast supports distribution data to group users, service needed for many potential DTN applications. While Internet mobile ad hoc has been studied extensively, due unique characteristic partitioning DTNs, DTNs is considerably...

Server replication is an approach often used to improve the ability of a service handle large number clients. One important factors in efficient utilization replicated servers direct client requests best server, according some optimality criteria. In this paper we target environment which are distributed across Internet, and clients identify using our application-layer any-casting service. Our goal allocate way that minimizes client's response time. To end, develop for estimating performance...

Parallel and distributed simulation tools are emerging that offer the ability to perform detailed, packet-level simulations of large-scale computer networks on an unprecedented scale. The state-of-the-art in network is characterized quantitatively. For this purpose, a metric based number packet transmissions can be processed by simulator per second wallclock time (PTS) used as means quantitatively assess performance. An approach realizing scalable leverages existing sequential models...

A new protocol for maintaining replicated data that can provide both high availability and low response time is presented. In the protocol, nodes are organized in a logical grid. Existing protocols designed primarily to achieve by updating large fraction of copies, which provides some (although not significant) load sharing. transaction processing shared effectively among storing copies data, experienced transactions system throughput improved significantly. The authors analyze use...

There is great potential for boosting the performance of mobile devices by offloading computation-intensive parts applications to cloud. The full realization this hindered a mismatch between how individual demand computing resources and cloud providers offer them: requests from device usually require quick response, may be infrequent, are subject variable network connectivity, whereas incur relatively long setup times, leased time quanta, indifferent connectivity. In paper, we present design...

Despite the increased capabilities of mobile devices, application resource requirements can often transcend what be accomplished on a single device. This has been addressed through several proposals for efficient computation offloading from devices to remote cloud resources or closely located computing known as cloudlets. In this paper we consider an environment in which computational is performed among set devices. We call Mobile Device Cloud (MDC). are interested MDCs where nodes {\em...

Network function virtualization (NFV) is a new network architecture framework that implements functions in software running on pool of shared commodity servers. NFV can provide the infrastructure flexibility and agility needed to successfully compete today's evolving communications landscape. Any service represented by chain (SFC) set VNFs be executed according given order. The needs instantiation VNF instances (VNFIs) are modules virtual machines. This paper deals with migration problem...

Discrete event simulation is widely used within the networking community for purposes such as demonstrating validity of network protocols and architectures. Depending on level detail modeled simulation, running time memory requirements can be excessive. The goal our research to develop demonstrate a practical, scalable approach parallel distributed that will enable widespread reuse sequential models software. We focus an parallelization where existing simulator build subnetworks are composed...