Minimal Communication Overhead Research Articles

RIDE: real-time massive image processing platform on distributed environment

As the demand for real-time data processing increases, a high-speed processing platform for large-scale stream data becomes necessary. For fast processing large-scale stream data, it is essential to use multiple distributed nodes. So far, there have been few studies on real-time massive image processing through efficient management and allocation of heterogeneous resources for various user-specified nodes on distributed environments. In this paper, we shall present a new platform called RIDE (Real-time massive Image processing platform on Distributed Environment) which efficiently allocates resources and executes load balancing according to the amount of stream data on distributed environments. It minimizes communication overhead by using a parallel processing strategy which handles the stream data considering both coarse-grained and fine-grained parallelism simultaneously. Coarse-grained parallelism is achieved by the automatic allocation of input streams onto partitions of broker buffer each processed by its corresponding worker node, and maximized by adaptive resource management which adjusts the number of worker nodes in a group according to the frame rate in real time. Fine-grained parallelism is achieved by parallel processing of task on each worker node and maximized by allocating heterogeneous resources such as GPU and embedded machines appropriately. Moreover, it provides a scheme of application topology which has a great advantage for higher performance by configuring the worker nodes of each stage using adaptive heterogeneous resource management. Finally, it supports dynamic fault tolerance for real-time image processing through the coordination between components in our system.

An enhanced un-split face-vertex flux-based VoF method

Multiple advancements of the dimensionally un-split, flux-based geometrical Volume-of-Fluid method on unstructured meshes are proposed. A second-order accurate interface reconstruction algorithm for both two-and-three dimensions is developed on unstructured hexahedral meshes. A new triangulation algorithm is developed that results in an absolute increase in accuracy and can be directly applied to other geometrical Volume-of-Fluid methods on both structured and unstructured meshes. A second-order accurate interpolation of point displacements is proposed for unstructured meshes. A global parallel error re-distribution algorithm is developed, with no negative effects on the L1 error convergence and computational costs. All algorithms are parallelized using the message passing parallel programming model, with a minimal parallel communication overhead. The results show L1 errors to be exactly numerically bounded, second-order convergent and lower than the errors reported for other contemporary methods, volume conservation that is near machine tolerance, as well as execution times comparable to those of methods developed on structured Cartesian meshes, regardless of the increased algorithmic complexity introduced by the connectivity of unstructured meshes.

Tiger hash based AdaBoost machine learning classifier for secured multicasting in mobile healthcare system

Secure multicast routing is an important in mobile healthcare system. Few research works have been developed to prevent malicious behaviors from disclosing integrity of data in mobile healthcare systems using machine learning technique. But, the performance of conventional machine learning technique was not effectual. In order to overcome this limitation, Tiger hashing based AdaBoost with SVM classifier (TH-ASVMC) technique is proposed. The TH-ASVMC technique is designed to improve the security of multicast routing in MANETs with higher data integrity rate and therefore reducing the time taken. The TH-ASVMC technique initially used Tiger hash function which converts the patient data to be transmitted over a wireless network into a hash value for maintaining the data integrity during the process of multicasting in mobile healthcare system. After that, TH-ASVMC technique used AdaBoost with SVM classifier to classify the nodes in mobile healthcare system as authentic or unauthentic based on measurement of trust value for securing multicast routing with minimum communication overhead. Thus, TH-ASVMC technique choose the only an authentic node for routing the hash value of patient data to multiple destination nodes in mobile healthcare system. This process results in enhanced reliability and scalability of secured multicast routing. The TH-ASVMC technique conducts the simulations works on metrics such as data integrity rate, scalability, reliability and communication overhead. The simulation results shows that the TH-ASVMC technique is able to improve the reliability and data integrity rate of multicast routing as compared to state-of-the-art works.

A comprehensive study of MPI parallelism in three-dimensional discrete element method (DEM) simulation of complex-shaped granular particles

A three-dimensional (3D) DEM code for simulating complex-shaped granular particles is parallelized using message-passing interface (MPI). The concepts of link-block, ghost/border layer, and migration layer are put forward for design of the parallel algorithm, and theoretical scalability function of 3-D DEM scalability and memory usage is derived. Many performance-critical implementation details are managed optimally to achieve high performance and scalability, such as: minimizing communication overhead, maintaining dynamic load balance, handling particle migrations across block borders, transmitting C++ dynamic objects of particles between MPI processes efficiently, eliminating redundant contact information between adjacent MPI processes. The code executes on multiple US Department of Defense (DoD) supercomputers and tests up to 2048 compute nodes for simulating 10 million three-axis ellipsoidal particles. Performance analyses of the code including speedup, efficiency, scalability, and granularity across five orders of magnitude of simulation scale (number of particles) are provided, and they demonstrate high speedup and excellent scalability. It is also discovered that communication time is a decreasing function of the number of compute nodes in strong scaling measurements. The code’s capability of simulating a large number of complex-shaped particles on modern supercomputers will be of value in both laboratory studies on micromechanical properties of granular materials and many realistic engineering applications involving granular materials.

PUF-Assisted Group Key Distribution Scheme for Software-Defined Wireless Sensor Networks

In this letter, a novel group key distribution scheme based on physical unclonable functions (PUFs) is proposed for software-defined wireless sensor networks. The PUF is difficult to be cloned and predicted, remarkably enhancing physical security performance of sensor nodes. In addition, under the centralized management of software-defined networking, PUF challenge is stored in the sensor nodes to minimize communication overhead. The proposed algorithm achieves group key delivery with two-way authentication function through one communication interaction. We demonstrate the scheme can resist multiple attacks, including cloning, eavesdropping, and so on, thus effectively guaranteeing the security and efficiency of group key distribution process.

Accurate Clock Discipline For Long-Term Synchronization Intervals

Efficient spectrum usage and optimized energy consumption directly depend on the radio duty cycle of the communicating devices. Tight synchronization of communicating nodes enables optimal orchestration of the access to the medium as nodes turn their radios on precisely when needed. Achieving synchronization, however, requires a reference time source to synchronize with and a periodic discipline mechanism to cope with the inherent clock drift. Packet-based synchronization or external time sources, such as GPS, are usually required to achieve that goal. When low-power operation is combined with bandwidth limitations and scale, as in the case of low power wide area networks, the overhead of such approaches is not affordable and advanced clock discipline mechanisms are required. In this paper, we propose a novel adaptive mechanism to discipline clocks in order to guarantee a 1 $ppm$ drift with minimal communication overhead.

Community-aware single-copy content forwarding in Mobile Social Network

One of the most challenging issues in Mobile Social Networks (MSNs) is to design a messages forwarding method that achieves high delivery and low communication overhead. Single-copy forwarding methods boasting minimum communication overhead are well-known in MSNs. However, attaining a reasonable delivery ratio by using a single-copy method is an open problem. A common way to resolve the problem is social-aware forwarding, but this faces two main weaknesses: one is that they either are unaware of community detection, or use supervised learning strategies, other is that they generally use the relay-destination contact probability for predicting future contacts without considering the contact time. In this paper, we propose community-aware forwarding (CAF) as a new single-copy forwarding method using Hidden Semi-Markov Model (HSMM) to find communities by utilizing the similarity of node contact patterns in different sojourn cycles. In this study, we train HSMM as an unsupervised algorithm to compute the node community transition and then propose a novel forwarding method that utilizes message expiration in order to make relay selection. Evaluation results confirm our superior CAF performance over the popular solutions investigated in terms of message delivery and latency.

An Acoustic Network Navigation System

This work describes a system for acoustic‐based navigation that relies on the addition of localization services to underwater networks. The localization capability has been added on top of an existing network, without imposing constraints on its structure/operation. The approach is based on the inclusion of timing information within acoustic messages through which it is possible to know the time of an acoustic transmission in relation to its reception. Exploiting such information at the network application level makes it possible to create an interrogation scheme similar to that of a long baseline. The advantage is that the nodes/autonomous underwater vehicles (AUVs) themselves become the transponders of a network baseline, and hence there is no need for dedicated instrumentation. The paper reports at sea results obtained from the COLLAB–NGAS14 experimental campaign. During the sea trial, the approach was implemented within an operational network in different configurations to support the navigation of the two Centre for Maritime Research and Experimentation Ocean Explorer (CMRE OEX) vehicles. The obtained results demonstrate that it is possible to support AUV navigation without constraining the network design and with a minimum communication overhead. Alternative solutions (e.g., synchronized clocks or two‐way‐travel‐time interrogations) might provide higher precision or accuracy, but they come at the cost of impacting on the network design and/or on the interrogation strategies. Results are discussed, and the performance achieved at sea demonstrates the viability to use the system in real, large‐scale operations involving multiple AUVs. These results represent a step toward location‐aware underwater networks that are able to provide node localization as a service.

Encode When Necessary

Recent research has shown that network coding has great potential to improve network performance in wireless communication. The performance of network coding in real-world scenarios, however, varies dramatically. It is reported that network coding brings negligible improvements but extra coding overhead in some scenarios. In this article, for the first time, we analyze the impact of link correlation on network coding and quantify the coding benefits. We propose correlated coding, which encodes packets only when performance improvement is achieved. Correlated coding uses only one-hop information, which makes it work in a fully distributed manner and introduces minimal communication overhead. The highlight of the design is its broad applicability and effectiveness. We implement the design with four broadcast protocols and three unicast protocols, and we evaluate them extensively on one 802.11 testbed and three 802.15.4 testbeds. The experimental results show that (i) more coding operations do not lead to fewer transmissions, and (ii) compared to existing network coding protocols, the number of transmissions is reduced with lower coding overhead.

Collaborative Topology Control for Many-to-One Communications in Wireless Sensor Networks

Topology control is relevant in wireless sensor network because of two reasons, namely minimal sensor coverage and power constraints. The former condition is typically satisfied by high-density deployment, whereas the latter mainly concerns with the control protocol design that is adaptable. Controlling communication topology is at the center of the efforts to optimize network performance while improving energy conservation. A dense topology often results in high interference and lower spatial reuse thus reduced capacity, while sparse topology is susceptible to network partitioning and sub-optimal path selection from the routing layer. Topology control has been extensively studied in both flat and hierarchical network by mean of power adjustment and clustering, respectively. Despite a common goal of making the topology less complex both techniques differ in their approach. While the focus of clustering is to form a connected backbone which consists of a minimum subset of nodes, i.e., dominating set, power adjustment focuses on minimizing energy consumption. Combining both approaches remains a relatively lesser explored area. We proposed a hybrid framework called collaborative topology control protocol, which combines dominating set-based clustering and transmission power adjustment. The protocol operates in two stages. During the first stage, a parameterized minimum virtual connected dominating set algorithm is executed to obtain clusters of various desirable properties. In the second stage, each cluster-head executes a distributed power adjustment algorithm. The simulation results show that the proposed topology control framework is capable of versatile performance in terms of transmission range/energy cost, the number of neighbors, edges, and hop distance. Moreover, the topology construction process uses the locally available information only with minimal communication overhead.

Service Oriented Cloud VM Placement Strategy for Internet of Things

The back-end intelligent system of Internet of Things (IoT) needs the powerful computing ability to support its service, and therefore, IoT is often constructed on the cloud platform. However, the resources for cloud computing are limited, and different types of services will have different demands for the resources. This paper proposed a service-oriented virtual machine (VM) placement strategy in cloud data center and divided the roles of VM into Web role, worker role, and storage according to the function types. On the basis of service orientation and in consideration of communication overhead between VMs in the data center, the genetic algorithm was used to conduct the optimal configuration for different types of VMs under the situation of limited resources to achieve the minimum communication overhead and total power consumption. The proposed cloud VM placement strategy is also suitable for the intelligent computing platform of IoT back end.

Distributed resource allocation in federated clouds

Cloud computing is an emerging technology which relies on virtualization techniques to achieve the elasticity of shared resources for providing on-demand services. When the service demand increases, more resources are required to satisfy the service demand. Single cloud generally cannot provide unlimited services with limited physical resources; therefore, the federation of multiple clouds may be one possible solution. In such environment, different cloud providers may own different pricing and resource allocating strategies. Thus, how to select the most appropriate provider to host applications becomes an important issue for clients. However, as the requests of accessing distributed resources increase, the occurrences of competing the same resource may also increase. In this study, a Distributed Resource Allocation (DRA) approach is proposed to solve resource competition in the federated cloud environment. Each job is supposed to consist of one or more tasks, and the communication behavior between tasks could be profiled. The proposed approach groups tasks according to communication behavior to minimize communication overhead, and tries to allocate grouped tasks to achieve equilibrium when resource competition occurs. Experimental results show that the cloud provider could obtain more profits by outsourcing resources in the federated cloud with enough resources.

Clustering in MANETs with Minimum Reclustering and Communication Overhead

Mobile Adhoc Networks (MANETs) are deployed quickly without the pre-existing infrastructure as other networks are created with proper infrastructure. Clustering offers hierarchical organization of mobile nodes by forming disjoint groups called clusters. The problem of scalability arises with increase in both mobility and the network size and in such cases clustering comes to rescue. Thus clustering, as a typical hierarchy, solves the issue of scalability in large dynamic MANETs, however it requires extra message exchange among mobile nodes for the cluster structure maintenance. The stability of cluster is greatly affected and it leads to the ripple effect of reclustering. This approach can eliminate the frozen period requirement and minimize the process of reclustering.

Fault-resilient localization for underwater sensor networks

Localization is one of the critical issues in underwater sensor networks (UWSNs) on grounds of environmental characteristics, mobility of sensor nodes, and lack of global positioning system (GPS) services. Mobility modeling and fault tolerance with respect to underwater characteristics are extremely difficult. In this energy constraint network, an adaptive localization scheme with minimum communication and GPS support is required. We propose a fault-resilient localization scheme, which provides good localization accuracy with minimal communication overhead. To resile from an anchor node failure, a sensor node predicts its position by learning the mobility behavior of its neighbors using multiple linear regression. Therefore, the data dissemination process can continue even after an unexpected case of anchor node failure. We simulate the network and analyze the localization accuracy. Several experiments with different test cases are conducted to analyze the validity of the location prediction system. Experimental results show that the last 10 location information from three immediate neighbors are adequate for accurate prediction and the localization scheme is well suited for medium-range UWSNs.

Energy efficient secure communication architecture for wireless sensor network

The wireless sensor network WSN is more vulnerable than the wired network because it is relatively easy for an adversary to eavesdrop, insert, alter, and intercept the message communicated in the network. Sensor nodes have very limited power and memory. While designing the secure communication architectures, the main goal is to design and memory efficient protocols. Most of the well-known communication architectures like TinySec, LLSP, and MiniSec compromise with the security issues because of and memory constraints of sensor nodes. In WSN, communication consumes more than computation. By minimizing communication overhead, the consumption can be reduced. We propose a new energy efficient secure communication called EESCA for WSN, to make the communication secure. Along with authentication, confidentiality and integrity, the proposed architecture provides strong replay protection, and data freshness as permutation and substitution are used to generate massage authentication code. EESCA is robust against heavy packet loss because reset or resynchronization of counter is not required. The proposed scheme is efficient as the security packet overhead is reduced by 1i¾źbyte over the best-known secure communication architectures. The efficacy of EESCA is shown through theoretical and experimental analysis. Copyright © 2016 John Wiley & Sons, Ltd.

CHARGING MANAGEMENT PROTOCOL FOR NEAR FIELD COMMUNICATION CHARGING

The current multiplicity of mobile communication devices has provided an impetus for the research into new mechanisms to supplement battery charge. Wireless charging is a solution that serves to eliminate the cable requirements of typical battery charging implementations. Numerous wireless charging implementations are based on inductive coupling, similar to existing non-radiative short range communication systems. This study proposes incorporating a charge management protocol into the existing Near Field Communication Interface and Protocol-1 (NFCIP-1) specification to achieve NFC-enabled wireless charging. To this end, the original NFCIP-1 protocol has been modified through a time-sharing arrangement to support a charging task within the protocol cycle. Simulations of the modified protocol cycle were implemented using an appropriate battery model and charging algorithm. Numerical results show that the modified protocol is able to charge the target battery with minimum communication overhead. Satisfactory performance is also observed for charging up to 2 target devices in a single session.

A Multi-GPU Fast Iterative Method for Eikonal Equations Using on-the-fly Adaptive Domain Decomposition

The recent research trend of Eikonal solver focuses on employing state-of-the-art parallel computing technology, such as GPUs. Even though there exists previous work on GPU-based parallel Eikonal solvers, only little research literature exists on the multi-GPU Eikonal solver due to its complication in data and work management. In this paper, we propose a novel on-the-fly, adaptive domain decomposition method for efficient implementation of the Block-based Fast Iterative Method on a multi-GPU system. The proposed method is based on dynamic domain decomposition so that the region to be processed by each GPU is determined on-the-fly when the solver is running. In addition, we propose an efficient domain assignment algorithm that minimizes communication overhead while maximizing load balancing between GPUs. The proposed method scales well, up to 6.17× for eight GPUs, and can handle large computing problems that do not fit to limited GPU memory. We assess the parallel efficiency and runtime performance of the proposed method on various distance computation examples using up to eight GPUs.

Threshold-Based Distributed Continuous Top-k Query Processing for Minimizing Communication Overhead

Threshold-Based Distributed Continuous Top-k Query Processing for Minimizing Communication Overhead

CorLayer: A Transparent Link Correlation Layer for Energy-Efficient Broadcast

Recent work has shown that wireless links are not independent, and that transmissions from a transmitter to multiple receivers are correlated. This finding has profound implications for the performance of network protocols such as broadcast, multicast, opportunistic routing, and network coding. In this paper, we show how link correlation can significantly impact broadcast. We present the design and implementation of CorLayer, a general supporting layer for energy-efficient reliable broadcast that carefully blacklists certain poorly correlated wireless links. The design uses only one-hop information, which makes it work in a fully distributed manner and introduces minimal communication overhead. The highlight of our work is CorLayer's broad applicability and effectiveness. We integrate CorLayer transparently with 16 state-of-the-art broadcast protocols specified in 13 publications on three physical testbeds running TelosB, MICAz, and GreenOrbs nodes, respectively. The experimental results show that CorLayer significantly improves energy efficiency across a wide spectrum of broadcast protocols and that the total number of packet transmissions can be reduced consistently by 47% on average.

Bayes Node Energy Polynomial Distribution to Improve Routing in Wireless Sensor Network.

Wireless Sensor Network monitor and control the physical world via large number of small, low-priced sensor nodes. Existing method on Wireless Sensor Network (WSN) presented sensed data communication through continuous data collection resulting in higher delay and energy consumption. To conquer the routing issue and reduce energy drain rate, Bayes Node Energy and Polynomial Distribution (BNEPD) technique is introduced with energy aware routing in the wireless sensor network. The Bayes Node Energy Distribution initially distributes the sensor nodes that detect an object of similar event (i.e., temperature, pressure, flow) into specific regions with the application of Bayes rule. The object detection of similar events is accomplished based on the bayes probabilities and is sent to the sink node resulting in minimizing the energy consumption. Next, the Polynomial Regression Function is applied to the target object of similar events considered for different sensors are combined. They are based on the minimum and maximum value of object events and are transferred to the sink node. Finally, the Poly Distribute algorithm effectively distributes the sensor nodes. The energy efficient routing path for each sensor nodes are created by data aggregation at the sink based on polynomial regression function which reduces the energy drain rate with minimum communication overhead. Experimental performance is evaluated using Dodgers Loop Sensor Data Set from UCI repository. Simulation results show that the proposed distribution algorithm significantly reduce the node energy drain rate and ensure fairness among different users reducing the communication overhead.