A Distributed Algorithm for Throughput Optimal Routing in Overlay Networks

Maximum throughput requires path diversity enabled by bifurcating traffic at different network nodes. In this work, we consider a network where traffic bifurcation is allowed only at a subset of nodes called routers, while the rest nodes (called forwarders) cannot bifurcate traffic and hence only forward packets on specified paths. This implements an overlay network of routers where each overlay link corresponds to a path in the physical network. We study dynamic routing implemented at the overlay. We develop a queue-based policy, which is shown to be maximally stable (throughput optimal) for a restricted class of network scenarios where overlay links do not correspond to overlapping physical paths. Simulation results show that our policy yields better delay over dynamic policies that allow bifurcation at all nodes, such as the backpressure policy. Additionally, we provide a heuristic extension of our proposed overlay routing scheme for the unrestricted class of networks.

The peer-to-peer (P2P) computing paradigm is an emerging paradigm that aims to overcome most of the main limitations of the traditional client/server architecture. In the P2P setting, individual computers communicate directly with each other in order to share information and resources without relying on any kind of centralized server. To achieve this full decentralization, an application-level (or overlay) network is constructed using, for example, TCP connections. In most of the existing P2P systems, the overlay network is built in a manner that does not guarantee that the overlay network is efficient with respect to a given metric (e.g. latency, hop count and bandwidth). Hence, an overlay node can be very far away, in terms of a given metric, from its overlay neighbors. This can result in both, an inefficient routing at the overlay network and an ineffective use of the underlying IP network. In this paper, we introduce a new measure, "goodness of overlay networks", to quantify the quality of an overlay network for a given metric. We then propose NetProber, a simple, distributed and scalable component that can be combined with any connected overlay network in order to allow the latter to adapt, and to become "good" within a finite amount of time.

It is crucial to use an efficient network-aware overlay network to enable Internet-scale multicast streaming service to adjust under the dynamic underlying network conditions and overlay node churn without extensive network measurements. We utilize our previously proposed network-aware SuperPeers-Peers geometric overlay network hierarchy which exploits the underlying network locality and proximity of the nodes, to build massive scale overlay multicast streaming network (Bos). We create source routed overlay multicast tree from the resulting optimal geometric routing paths based on the shortest geometric distances between SuperPeers at the SuperPeers layer. The overlay multicast tree is self-healed locally in a distributed manner using an ancestor-recovery protocol. The high bandwidth multicast backbone infrastructure at the SuperPeers layer will serve the Peers in the lower layer. We evaluate Bos on ten massive scale networks each consisting of 100,000 nodes and in the PlanetLab. Our massive scale overlay multicast performance results show high efficiency and good scalability.

Inter-domain policy violations in multi-hop overlay routes: Analysis and mitigation

We present a new routing policy, called maximum shortest paths (MP) routing policy, within the class of shortest-path routing policies for mesh-connected topologies which include popular 2-D and 3-D meshes, 2-D and 3-D tori, and n-dimensional hypercubes (n-cubes). In this policy, the routing message is always forwarded to a neighbor from which there exists a maximum number of shortest paths to the destination. An optimal routing defined in this paper is the one that maximizes the probability of reaching the destination from a given source without delays at intermediate nodes. We show that the MP routing policy is equivalent to the e-cube routing in n-cubes which is optimal, and it is also equivalent to the Badr and Podar's zig-zag (Z/sup 2/) routing policy in 2-D meshes which is also optimal. We prove that the Z/sup 2/ routing policy is not optimal in any N/spl times/N torus, where N is an even number larger than four. A routing algorithm is proposed to implement the MP routing policy in 2-D tori and if is proved to be at least suboptimal (optimal for some cases). Our approach is the first attempt to address optimal routing in the torus network which is still an open problem.

Shipping in ice covered regions has gained high attention within recent years. Analogous to weather routing, the occurrence of ice in a seaway affects the selection of the optimal route with respect to the travel time or fuel consumption. The shorter, direct path between two points — which may lead through an ice covered area — may require a reduction of speed and an increase in fuel consumption. A longer, indirect route, could be more efficient by avoiding the ice covered region. Certain regions may have to be avoided completely, if the ice thickness exceeds the ice-capability of the ship. The objective of this study is to develop a computational method that combines coastline maps, route cost information (e.g. ice thickness), transport task and ship properties to find the optimal route between port of departure A and port of destination B. The development approach for this tool is to formulate the transport task in form of a potential problem, solve this equation with a finite element method and apply edge detection methods and line integration to determine the optimal route. The functionality of the method is first evaluated with simple test problems and then applied to realistic transport scenarios.

Overlay networks have emerged as a powerful and flexible platform for developing new disruptive network applications. The performance and reliability of overlay applications depend on the capability of overlay networks to dynamically adapt to various factors such as link/node failures, overlay link quality, and overlay node characteristics. In order to achieve this, the overlay applications require scalable and open overlay monitoring services to monitor, aggregate globally distributed events and take appropriate control actions. In this paper, we propose the techniques and algorithms to create an optimal event monitoring and aggregation infrastructure (called MOON) that minimizes the monitoring latency (i.e., event retrival/detection time) and event aggregation cost (i.e., intrusiveness) considering the large-scale geographical and network distribution of overlay nodes. The proposed monitoring infrastructure, MOON, clusters and organizes overlay nodes efficiently such that overlay applications can globally monitor and query correlated events in an overlay network with minimum latency and monitoring cost. Our simulations and experimental studies show the evaluation of MOON under many various topological structures, network sizes, and event aggregation volumes.

The network routing problem is one of important problems to consider behaviors of multiple flows on a network with traffic. Some network routing problems can be classified into the selfish routing game in which the behaviors of the flows are competitive and the optimal routing problem in which the behaviors of the flows are cooperative. This study focuses on the optimal routing problem. The optimal routing problem is formulated as an optimization problem with a linear equality constraint and lower limit constraints. The learning problem of the support vector machine (SVM) is formulated as an optimization problem with a linear equality constraint and upper and lower limit constraints. The sequential minimal optimization (SMO) is one of the fast solvers for the learning problem. SMO can be applied to solve the optimal routing problem. This study proposes a new fast optimal routing search method using SMO. In the numerical experiments, the effectiveness of the proposed method on the computational cost is verified with the comparison to the search method using the replicator dynamics.

We introduce dynamic internal message routing algorithm in overlay network used by distributed file system. Messages containing system requests are usually smaller than messages carrying data, but their number is much higher. Each kind of message has special requirements on quality of service. Messages carrying data require high bandwidth. Messages containing system requests are crucial for system function. They manage system operations, like replication, synchronization, locking, and metadata transfer exchange, and consequently, need as low latency as possible. We describe design and implementation of routing algorithm, which takes into account different quality of service requirements based on kind of message. Messages are classified by quality of service requirements. Class of the message determines the chosen route. Proposed concept has been proved on experimental distributed file system developed in the Department. We have verified that optimization of dynamic routing of system request has positive impact on system latency and throughput.

When dealing with multimedia delivery, scalability, efficiency and Quality of Service (QoS) are challenging features for today's networks. Overlay networks can compensate for the lack of control over the network routers, but for real-world applications overlay network performance is an issue. This paper presents an efficient solution for software implementation of overlay network nodes. A new, advanced Distributed Linked Data Structures (DLDS) schema has been developed to provide scalable per-flow resource reservation and fast, searchless routing/forwarding table lookup. This schema has been integrated in a new overlay network framework that features hierarchical dynamic routing and QoS estimation based on performance measures of the underlying network. The proposed technique has been implemented and tested in a large, realistic network simulation environment, and experimental results are presented.

Optimizing resource allocation and routing to satisfy service needs is paramount in large-scale networks. Software-defined networking (SDN) is a new network paradigm that decouples forwarding and control, enabling dynamic management and configuration through programming, which provides the possibility for deploying intelligent control algorithms (such as deep reinforcement learning algorithms) to solve network routing optimization problems in the network. Although these intelligent-based network routing optimization schemes can capture network state characteristics, they are prone to falling into local optima, resulting in poor convergence performance. In order to address this issue, this paper proposes an African Vulture Routing Optimization (AVRO) algorithm for achieving SDN routing optimization. AVRO is based on the African Vulture Optimization Algorithm (AVOA), a population-based metaheuristic intelligent optimization algorithm with global optimization ability and fast convergence speed advantages. First, we improve the population initialization method of the AVOA algorithm according to the characteristics of the network routing problem to enhance the algorithm’s perception capability towards network topology. Subsequently, we add an optimization phase to strengthen the development of the AVOA algorithm and achieve stable convergence effects. Finally, we model the network environment, define the network optimization objective, and perform comparative experiments with the baseline algorithms. The experimental results demonstrate that the routing algorithm has better network awareness, with a performance improvement of 16.9% compared to deep reinforcement learning algorithms and 71.8% compared to traditional routing schemes.

Service Overlay Networks (SONs) allow virtual operators to create and deploy value-added Internet services with Quality of Service guarantees, while leaving the underlying network infrastructure unchanged. The deployment of a SON can be very expensive, and hence its planning requires careful decisions, including the overlay nodes’ placement and the capacity provisioning of the access links that connect the end-users to the SON infrastructure.In this work we first propose a novel Integer Linear Programming (ILP) model for the overlay network design problem which selects the optimal number and position of overlay nodes, the capacity reserved on each overlay link, as well as the optimal routing of the incoming traffic demands.Since such model can be solved to the optimum only for small network instances, we further propose an efficient and novel tabu search based heuristic for the planning of SONs that combines polynomial size and very large-scale neighborhoods. The very large-scale neighborhood of the solution given by tabu search is explored efficiently to obtain in a short time a new one that is both far from the current solution and cost-decreasing.We provide numerical results of the proposed heuristic on a set of realistic, large-size instances, including real ISP topologies, and discuss the effect of different parameters on the characteristics of the planned networks. Furthermore, we compare such results with the bound obtained solving our ILP model in small network scenarios. We show that in the considered network topologies the proposed heuristic performs very close to the optimum with a short computation time, thus providing a promising framework for the design of SONs.

Distributed hash table (DHT)-based overlay networks, represented by Pastry, CAN, and Chord, offer an administration-free and fault-tolerant application-level overlay network. While elegant from a theoretical perspective, these systems have some disadvantages. First, they rely on application-level routing, which may be inefficient with respect to network delays and bandwidth consumption. Second, they typically construct a homogeneously structured overlay even though nodes in these networks usually have varying physical connectivity and packet-forwarding capacities. In this paper, we propose two approaches for constructing an auxiliary expressway network to take advantage of the different connectivity, forwarding capacities, and availabilities of the nodes. As a result, we are able to reconcile the conflict of presenting the applications with a homogeneous structured overlay to simplify management, while at the same time taking advantage of the inherent heterogeneity of the underlying physical network to speed up routing. Our simulation results show that our expressway can achieve close to optimal routing performance (on average, 1.07 and 1.41 times optimal routing for an Internet-like topology and a large synthesized transit-stub graph, respectively) in overlay networks.

Determining an accurate and reliable prediction of the optimal routing paths between source and destination nodes in Mobile Adhoc Network (MANET) is always challenge due to dynamic nature of mobile nodes and the scarcity of network availability. The purpose of this work is to identify reliable optimal paths (ordered) from the set of possible random routing paths. The proposed model leverages fuzzy sets and rough sets for routing path classification. It also utilizes TOPSIS (Technique for order of performance by similarity to ideal solution), a multi criteria desicion analysis method for obtaining best and worst routes in the network. The derived routing path overcomes dynamic traffic routing problem in MANET. The proposed model compared with existing optimal routing models (Genetic Algorithm, Anti-Colony Optimization Algorithm) and found that CPU computing time decreased by 5X and through put (TP) increased by 2X.

In the context of modern high-speed Internet network, routing is often complicated by the notion of guaranteed quality of service (QoS), which can either be related to time, packet loss or bandwidth requirements: constraints related to various types of QoS make some routing inacceptable. Due to emerging real-time and multimedia applications, efficient routing of information packets in dynamically changing communication network requires that as the load levels, traffic patterns and topology of the network change, the routing policy also adapts. We focused in this paper on QoS based routing by developing a neuro-dynamic programming to construct dynamic state-dependent routing policies. In this paper, we propose an approach based on adaptive algorithm for packet routing using reinforcement learning called N best optimal path Q routing algorithm (NOQRA) which optimizes two criteria: cumulative cost path (or hop count if each link cost =1) and end-to-end delay. A load balancing policy depending on a dynamical traffic path probability distribution function is also defined and embodied in NOQRA to characterize the distribution of the traffic over the N Best Paths. Numerical results obtained with OPNET simulator for different levels of traffic's load show that NOQRA gives better results compared to standard optimal path routing and Q-routing algorithm based on Q-learning paradigm.