Network Partitioning Scheme Research Articles

Optimization method of skeleton network partitioning scheme considering resilience active improvement in power system restoration after typhoon passes through

After a major blackout caused by typhoon, secondary meteorological or geological disasters may also occur, so the secondary fault probability of transmission components in the restoration process is considerably higher than that in the normal operation state. A robust and reliable skeleton network restoration scheme is important to prevent the recurrence of collapse during restoration. Secondary component faults or contingencies are rarely considered in the existing research on transmission system restoration. Given that, this paper proposes a network partitioning scheme considering active resilience improvement of the skeleton network in parallel system restoration after a typhoon passes through. First, the necessity of considering secondary faults in the restoration after the blackout caused by typhoon is analyzed, and a new resilience index considering the impact of secondary faults is proposed. Next, the optimization objectives are proposed, including maximizing system resilience and minimizing restoration operation cost, and the basic partitioning constraints were modeled. Linearized operation constraints considering secondary faults were proposed to consider the resilience index in the partitioning scheme optimization. Then, a multi-objective mixed integer linear programming (MILP) network partitioning model considering the active resilience improvement of the skeleton network was established to mitigate the effect of secondary faults caused by typhoon. Furthermore, the Normalized Normal Constraint (NNC) method and the Variation Coefficient (VC) method were used to solve the multi-objective MILP model. To reduce the scale of the decision space, a swing-bus-based solving simplification strategy was proposed. In this paper, New England 10-unit 39-bus system and IEEE 118-bus system are taken as cases to demonstrate the proposed partitioning scheme optimization method considering the system resilience improvement. The simulation results show that the partitioning schemes obtained can effectively reduce the load loss of the system under the influence of secondary faults, and lead to a more reliable and smooth restoration process of transmission network after the typhoon passes through.

Adaptive deep-learning equalizer based on constellation partitioning scheme with reduced computational complexity in UVLC system.

Visible light communication (VLC) system has emerged as a promising solution for high-speed underwater data transmission. To tackle with the linear and nonlinear impairments, deep learning inspired equalization is introduced into VLC. Despite their success in accuracy, deep learning approaches often come with high computational budget. In this paper, we propose an adaptive deep-learning equalizer based on complex-valued neural network and constellation partitioning scheme for 64 QAM-CAP modulated underwater VLC (UVLC) system. Inspired by the fact that symbols modulated at different levels experience various extent of nonlinear distortion, we adaptively partition the received symbols in constellation and design compact equalization networks for specific regions to reduce computation consumption. Experiments demonstrate that the partitioned equalizer can achieve the bit error rate below the 7% hard-decision forward error correction (HD-FEC) limit of 3.8 × 10-3 at 2.85 Gbps similar to the standard complex-valued network, yet with 56.1% total computational complexity reduction. This work paves the path for online data processing in high speed UVLC system.

FSA: A Fine-Grained Systolic Accelerator for Sparse CNNs

Sparsity, as an intrinsic property of convolutional neural networks (CNNs), has been widely employed for hardware acceleration, and many customized accelerators tailored for sparse weights or activations have been proposed in these years. However, the irregular sparse patterns introduced by both weights and activations are much more challenging for efficient computation. For example, due to the issues of access contention, workload imbalance, and tile fragmentation, the state-of-the-art sparse accelerator SCNN fails to fully leverage the benefits of sparsity, leading to nonoptimal results for both speedup and energy efficiency. In this article, we propose an efficient sparse CNN accelerator for both weights and activations, namely finegrained systolic accelerator (FSA), which jointly optimizes both hardware dataflow and software partitioning and scheduling strategy. Specifically, to deal with the access contentions problem, we present a fine-grained systolic dataflow, in which the activations move rhythmically along the horizontal processing element array while the weights are fed into the array in a fine-grained order. We then propose a hybrid network partitioning strategy that sets different partitioning strategies for different layers to balance the workload and alleviate the fragmentation problem caused by both sparse weights and activations. Finally, we present a scheduling search strategy to find the optimized schedules for neural networks, which can further improve energy efficiency. Extensive evaluations show that the proposed FSA consistently outperforms SCNN over AlexNet, VGGNet, GoogLeNet, and ResNet with an average speedup of 1.74× and up to 13.86× energy efficiency.

Nonparametric Significance Test for Discovery of Network‐Constrained Spatial Co‐Location Patterns

Spatial co‐location patterns are useful for understanding positive spatial interactions among different geographical phenomena. Existing methods for detecting spatial co‐location patterns are mostly developed based on planar space assumption; however, geographical phenomena related to human activities are strongly constrained by road networks. Although these methods can be simply modified to consider the constraints of networks by using the network distance or network partitioning scheme, user‐specified parameters or priori assumptions for determining prevalent co‐location patterns are still subjective. As a result, some co‐location patterns may be wrongly reported or omitted. Therefore, a nonparametric significance test without priori assumptions about the distributions of the spatial features is proposed in this article. Both point‐dependent and location‐dependent network‐constrained summary statistics are first utilized to model the distribution characteristics of the spatial features. Then, by using these summary statistics, a network‐constrained pattern reconstruction method is developed to construct the null model of the test, and the prevalence degree of co‐location patterns is modeled as the significance level. The significance test is evaluated using the facility points‐of‐interest data sets. Experiments and comparisons show that the significance test can effectively detect network‐constrained spatial co‐location patterns with less priori knowledge and outperforms two state‐of‐the‐art methods in excluding spurious patterns.

On reliability improvement of Software-Defined Networks

In Software-Defined Networks (SDNs) the role of the centralized controller is crucial, and thus it becomes a single point of failure. In this work, a distributed controller architecture is explored as a possible solution to improve fault tolerance. A network partitioning strategy, with small subnetworks, each with its own Master controller, is combined with the use of Slave controllers for recovery aims. A novel formula is proposed to calculate the reliability rate of each subnetwork, based on the load and considering the number and degree of the nodes as well as the loss rate of the links. The reliability rates are shared among the controllers through a newly-designed East/West bound interface, to select the coordinator for the whole network. This proposed method is called “Reliable Distributed SDN (RDSDN).” In RDSDN, the failure of controllers is detected by the coordinator that may undertake a fast recovery scheme to replace them. The numerical results prove performance improvement achievable with the adoption of the RDSDN and show that this approach performs better regarding failure recovery compared to methods used in related research.

Network partitioning strategy for parallel power system restoration

Parallel restoration is an efficient way to speed up the restoration process after a wide-area outage or a complete blackout of a power system. To implement efficient parallel restoration, an appropriate network partitioning method is necessary. Given this background, a two-step network partitioning strategy for parallel power system restoration is presented. In order to minimise the restoration time of generating units, a grouping model for non-black-start units is developed first. Based on the graph theory, a power network partitioning model is presented with the objectives of minimising the number of the interconnected lines and maximising the electrical distance of the interconnected lines among different restoration subsystems. The well-established AMPL/CPLEX is employed to solve the developed optimisation models. Finally, the IEEE New England 10-unit 39-bus system and a part of Zhejiang provincial power system in China are employed to illustrate the feasibility and effectiveness of the developed models.

Decentralized voltage control of clustered active distribution network by means of energy storage systems

The paper presents a network partitioning strategy for the optimal voltage control of Active Distribution Networks (ADNs) actuated by means of a limited number of Distributed Energy Storage Systems (DESSs). The proposed partitioning uses a linear programming approach by means of the known concept of voltage sensitivities. Then, two decentralized optimal control algorithms are proposed relying, respectively, on the Thévenin equivalents and a recursive approach. These algorithms are developed using the Multi-Agent System (MAS) concept. With respect to a centralized control algorithm, the aim of the network clustering is to reduce the number of exchanged messages among the clusters when one of the two proposed decentralized control algorithms is adopted. The effectiveness of the two proposed controls is assessed with respect to the performances of the equivalent centralized control using numerical examples composed by the IEEE 13 and IEEE 123 buses distribution test feeders adapted to include stochastic generation and DESSs.

Analysis of Base Station Assisted Novel Network Design Space for Edge-based WSNs

Limited and constrained energy resources of wireless sensor network should be used wisely to prolong sensor nodes lifetime. To achieve high energy efficiency and to increase wireless sensor network lifetime, sensor nodes are grouped together to form clusters. Organizing wireless sensor networks into clusters enables the efficient utilization of limited energy resources of the deployed sensor nodes. However, the problems of unbalanced energy consumption exist in intra and inter cluster communication, and it is tightly bound to the role and the location of a sensor nodes and cluster heads in the network. Also, clustering mechanism results in an unequal load distribution in the network. This paper presents an analytical and conceptual model of Energy- efficient edge-based network partitioning scheme proposed for wireless sensor networks. Also, it analyzes different network design space proposed for wireless sensor networks and evaluates their performance. From the experimental results it is observed that, with proper network organization mechanism, sensor network resources are utilized effectively to elevate network lifetime.

Zone-Based Routing Protocol for Wireless Sensor Networks.

Extensive research happening across the globe witnessed the importance of Wireless Sensor Network in the present day application world. In the recent past, various routing algorithms have been proposed to elevate WSN network lifetime. Clustering mechanism is highly successful in conserving energy resources for network activities and has become promising field for researches. However, the problem of unbalanced energy consumption is still open because the cluster head activities are tightly coupled with role and location of a particular node in the network. Several unequal clustering algorithms are proposed to solve this wireless sensor network multihop hot spot problem. Current unequal clustering mechanisms consider only intra- and intercluster communication cost. Proper organization of wireless sensor network into clusters enables efficient utilization of limited resources and enhances lifetime of deployed sensor nodes. This paper considers a novel network organization scheme, energy-efficient edge-based network partitioning scheme, to organize sensor nodes into clusters of equal size. Also, it proposes a cluster-based routing algorithm, called zone-based routing protocol (ZBRP), for elevating sensor network lifetime. Experimental results show that ZBRP out-performs interims of network lifetime and energy conservation with its uniform energy consumption among the cluster heads.

A flexible black-start network partitioning strategy considering subsystem recovery time balance

Summary When black out occurs, the whole power grid is divided into several subsystems to increase the efficiency and shorten the duration of power system restoration by using parallel restoration strategy. Therefore, reasonable network partitioning in black start phase after large-scale power failures is important in both theory and practice. In this paper, a flexible black start network partitioning strategy considering the subsystem recovery time balance is proposed. Based on the rules of zone division and analyzing initial black start partitioning model of subsystem shortest recovery path, subsystem recovery time balance index is defined to represent disparity of every subsystem's restoration process. Meanwhile, a flexible factor of total recovery time is introduced as part of objective function with the time balance index, aiming at reducing waiting time of the synchronization operation and strengthens efficiency of black start at the cost of the least total time increment. The classic Dijkstra algorithm and improved genetic algorithm are employed to obtain the optimal restoration plan in the paper. Case study on IEEE 118-bus test system shows the validity of the methodology proposed. Copyright © 2014 John Wiley & Sons, Ltd.

A speedup technique for dynamic graphs using partitioning strategy and multithreaded approach

There are many pre-processing-based speedup techniques for shortest path problems that are available in the literature. These techniques have an increased demand because of large datasets in such applications such as roadmaps, web search engines and mobile data sets. Pre-processing for the Time-Dependent Shortest Path Problem is still a demanding process that involves graph or network partitioning strategy. Efficient pre-processing of graphs or networks reduces the shortest path computation time while parallelizing the pre-processing phase improves the speedup of the system. In this paper, a speedup technique called Recursive Spectral Bisection (RSB) combined with the Elliptic Convolution of the shortest path method is proposed for dynamic Time-Dependent networks. The same method has been parallelized, and the results are tested on three types of graphs. It is observed that the Time-Dependent RSB combined with the Elliptic Convolution of the shortest path method has no update time, and the Query Performance Loss (QPL) is reduced in planar and road networks compared to random networks. In road networks, the proposed method achieves an average speedup in a QPL of 140. The use of the Parallel speedup technique results in an average speedup in a QPL of more than 1 in the planar and road networks.

TM: a new and simple topology for interconnection networks

The selection of a topology is essential to the performance of interconnection networks, so designing a new, cost-effective topology is very significant. 2D mesh is one of the most popular topologies. However, the diameter and average distance of a 2D mesh are large enough to greatly influence the performance of the network. This paper presents a novel topology called TM, which combines the advantages of both a 2D torus and a 2D mesh. For an n×n network, the total number of links in a TM is the same as that in a mesh, while the diameter of a TM is extremely close to that of a torus. Besides, the average distance of a TM is at the middle of that of a torus and that of a mesh. To prevent deadlocks in TMs, a virtual network partitioning scheme is adopted into the TM network. Moreover, both of the deterministic and fully-adaptive routing techniques in TMs are proposed in this paper. Compared to mesh, the TM network provides average distance and diameter reduction, which contributes to the performance enhancement. Sufficient simulation results are presented to show the effectiveness of the TM network, and the new routing schemes proposed for it, by comparing with the mesh network. Compared to the torus, which requires at least 3 virtual channels to support fully-adaptive routing, the TM network can support fully-adaptive routing with only 2 virtual channels. Seen from the experimental results, in most cases, the performance of TM is worse than the torus, while in some cases, the performance of TM is comparable to torus or even better than the torus.

Deadlock-Free Adaptive Routing in Meshes with Fault-Tolerance Ability Based on Channel Overlapping

A new deadlock-free routing scheme for meshes is proposed based on a new virtual network partitioning scheme, called channel overlapping. Two virtual networks can share some common virtual channels based on the new virtual network partitioning scheme. The deadlock-free adaptive routing method is then extended to deadlock-free adaptive fault-tolerant routing in 3D meshes still with two virtual channels. A few faulty nodes can make a higher dimensional mesh unsafe for fault-tolerant routing methods based on the block fault model, where the whole system (n-dimensional space) forms a fault block. Planar safety information in meshes is proposed to guide fault-tolerant routing and classifies fault-free nodes inside 2D planes. Many nodes globally marked as unsafe in the whole system become locally enabled inside 2D planes. This fault-tolerant deadlock-free adaptive routing algorithm is also extended to the one in an n-dimensional meshes with two virtual channels. Extensive simulation results are presented and compared to previous methods.

COOPERATIVE SELF-COMPOSITION AND DISCOVERY OF GRID SERVICES IN P2P NETWORKS

The desirable global scalability of Grid systems has steered the research towards the employment of the peer-to-peer (P2P) paradigm for the development of new resource discovery systems. As Grid systems mature, the requirements for such a mechanism have grown from simply locating the desired service to compose more than one service to achieve a goal. In Semantic Grid, resource discovery systems should also be able to automatically construct any desired service if it is not already present in the system, by using other, already existing services. In this paper, we present a novel system for the automatic discovery and composition of services, based on the P2P paradigm, having in mind (but not limited to) a Grid environment for the application. The paper improves composition and discovery by exploiting a novel network partitioning scheme for the decoupling of services that belong to different domains and an ant-inspired algorithm that places co-used services in neighbouring peers.

A dual-hamiltonian-path-based multicasting strategy for wormhole-routed star graph interconnection networks

Multicast is an important collective communication operation on multicomputer systems, in which the same message is delivered from a source node to an arbitrary number of destination nodes. The star graph interconnection network has been recognized as an attractive alternative to the popular hypercube network. In this paper, we first address a dual-hamiltonian-path-based routing model with two virtual channels based on two hamiltonian paths (HPs) and a network partitioning strategy for wormhole-routed star graph networks. Then, we propose three efficient multicast routing schemes on basis of such a model. All of the three proposed schemes are proved deadlock-free. The first scheme, network-selection-based dual-path routing, selects subnetworks that are constructed either by the first HP or by the second HP for dual-path routing. The second one, optimum dual-path routing, selects subnetworks with optimum routing path for dual-path routing. The third scheme, two-phase optimum dual-path routing, includes two phases, source-to-relay and relay-to-destination. Finally, experimental results are given to show that our proposed three routing schemes outperform the unicast-based, the HP, and the single-HP-based dual-path routing schemes significantly.

A scalable parallel formulation of the backpropagation algorithm for hypercubes and related architectures

We present a new technique for mapping the backpropagation algorithm on hypercube and related architectures. A key component of this technique is a network partitioning scheme called checkerboarding. Checkerboarding allows us to replace the all-to-all broadcast operation performed by the commonly used vertical network partitioning scheme, with operations that are much faster on the hypercubes and related architectures. Checkerboarding can be combined with the pattern partitioning technique to form a hybrid scheme that performs better than either one of these schemes. Theoretical analysis and experimental results on nCUBE and CM5 show that our scheme performs better than the other schemes, for both uniform and nonuniform networks.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>