Search

The emerging real-time hyper-physical system (CPS), such as autonomous vehicle and live interactive media application, requires time deterministic behaviour. This is challenging to achieve by using the traditional general purpose operating system (GPOS). This paper presents a new design of the real-time operating system (OS) scheduling mechanism called 'time deterministic cyclic scheduling' (TDCS) mainly for live multimedia tasks processing. This new scheduler shares a similar philosophy as classic cyclic execution but with flexibility and dynamic configuration. This hybrid design is based on both time-reserved based cyclic execution and priority-based pre-emptive scheduling for mixed criticality applications. The simulation results show that this scheduling scheme can achieve predictable timing behaviour of task delay and jitter under high CPU utilisation. This shows that the proposed scheme is promising for low latency high-performance multimedia censoring tasks that occur in a periodic manner. [ABSTRACT FROM AUTHOR]

With the exponential growth of network data storage scale, the issue of uniform distribution and efficient retrieval of data in the distributed storage systems such as the Redis cluster has received increasing attention in recent years. In view of the existing problems in scalability, usability and other aspects of the solution in current researches, we propose the distributed dynamic cuckoo filter system based on Redis cluster. On one hand, we introduce an efficient hash indexing structure–dynamic cuckoo filter (DCF), which only stores the fingerprint information of data, and has the automatically scalable capacity to meet the demand of data storage on a dynamic scale. On the other hand, we use an improved consistent hashing algorithm to construct Redis cluster and use the thorough communication mechanism of Redis cluster to achieve the data sharing and efficient utilisation of multi-machine filters. The scheme proposed in this paper can take the time and space efficiency into account, greatly promote the retrieval performance of massive data, and improve the reliability and availability of Redis cluster. [ABSTRACT FROM AUTHOR]

We give a survey on the history of computers in Japan that have been produced over the last 50 years. Here, we focus on the topics of personal computers (PCs). We can see PCs produced in Japan have very unique features. They coevolved with some other related machines, e.g. word processing machines, game machines, etc. They also had several localised features such as capability of dealing with Japanese language. At the same time, developers of PCs struggled to find a good solution between localisation and globalisation. In the early stage of the history of PCs, their architectures were all different, and software had to be developed and tuned for each of them. Thus providing a common platform to all PCs on which every software can run was indispensable for globalisation. In this paper, how PCs in Japan have been evolved and how such problems were attacked are discussed based on the author's experiences of developing early PCs. [ABSTRACT FROM AUTHOR]

This paper tells my recollections of half a century of computing in Canada and, briefly earlier, in two other corners of the world. [ABSTRACT FROM AUTHOR]

The significant increase in the number of processors of multiprocessor system increases its vulnerability to component failures. Diagnosability is an important indicator for the reliability of interconnection networks. The g-good-neighbour conditional diagnosability is the maximum number of faulty vertices a network can guarantee to identify, under the condition that every fault-free vertex has at least g fault-free neighbours (i.e. good neighbours). In this paper, we establish that the 1-good-neighbour conditional diagnosability of (n , k) -enhanced hypercube network Q n , k under PMC model is 2n−1 for n = k + 1 and k ≥ 3 , or 2n + 1 for n>k + 1 and k ≥ 3 , respectively. Moreover, the 2-good-neighbour conditional diagnosability of Q n , k is 3n−3 for n = k + 1 and k ≥ 4 , or 4n−5 for n = k + 2 and k ≥ 3 , or 4n−1 for n>k + 2 and k ≥ 3 , respectively. [ABSTRACT FROM AUTHOR]

A cyclic vertex-cut of a graph G is a vertex set S such that G−S is disconnected and at least two of its components contain cycles. If G has a cyclic vertex-cut, then it is said to be cyclically separable. For a cyclically separable graph G, the cyclic vertex-connectivity κ c (G) is defined as the cardinality of a minimum cyclic vertex-cut. Let G i be a k i -regular (k i ≥ 2) and maximally connected graph with girth g (G i) ≥ 5 for i=1,2. In this paper, we mainly prove that κ c (K m ◻ G 2) = 3 k 2 + m − 3 for m ≥ 3 and κ c (G 1 ◻ G 2) = 4 k 1 + 4 k 2 − 8. In addition, we state sufficient conditions to guarantee κ c (K 2 ◻ G 2) = 2 κ (G 2). [ABSTRACT FROM AUTHOR]

Fault diagnostic analysis is extremely important for interconnection networks. Given a graph G and a positive integer g, the g-extra connectivity of G (denoted by κ o (g) (G)) is the minimum cardinality of a subset S of V (G) such that G−S is disconnected and every remaining component has at least g + 1 vertices. The g-extra diagnosability of G (denoted by t ~ g (G)), is the maximum number of faulty vertices that the system can guarantee to identify under the condition that every fault-free component contains at least g + 1 vertices. The t/k-diagnosis strategy can detect up to t faulty vertices which might include at most k misdiagnosed vertices. In this paper, we first determine κ o (g) (C C N (n)) = (g + 1) n − g + 1 2 + 1 for n ≥ 2 , 1 ≤ g ≤ n − 1 , where C C N (n) is an n-dimensional complete cubic network, which generalises the hierarchical cubic network. Moreover, we establish t ~ g (C C N (n)) = (g + 1) (n + 1) − g + 1 2 under the PMC model ( n ≥ 4 , 1 ≤ g ≤ n − 3) and under the MM* model ( n ≥ 6 , 1 ≤ g ≤ (n − 2) / 4 ), respectively. Furthermore, we show that C C N (n) is [ (k + 1) n − k + 1 2 + 1 ] / k -diagnosable under the PMC model. As a consequence, we also derive the related results of the n-dimensional hierarchical cubic network H C N n . [ABSTRACT FROM AUTHOR]

The k-ary n-cube is a popular interconnection network for many multiprocessor systems because of its good topological structure. In this paper, we investigated the relationship between the fault diagnosability of k-ary n-cube with missing edges and its minimum degree of vertices. Let G be a k-ary n-cube with missing edges and the minimum degree δ (G) = r. Then we show that the diagnosability of G is r for r ≥ 3 , n ≥ 2 under the PMC model and BGM model. [ABSTRACT FROM AUTHOR]

The diagnosability of multiprocessor systems is an important research topic. In 2017, Wang et al. extended the k-ary n-cube and defined an expanded k-ary n-cube, which is a system topology structure with good properties. In this paper, we prove that the expanded k-ary n-cube has the strong local diagnosability property for n ≥ 2 and k ≥ 6 , and it keeps this strong property even if there exist (4 n − 5) missing edges in it under the comparison model. [ABSTRACT FROM AUTHOR]

The diagnosability of a multiprocessor system plays an important role. The system and an interconnection network have a underlying topology, which are usually presented by a graph. The exchanged hypercube E H (s , t) has some good properties. In this paper, we study the diagnosis on E H (s , t) under the comparison model. Following the concept of the local diagnosability, the strong local diagnosability property is discussed. This property describes the equivalence of the local diagnosability of a node and its degree. We prove that E H (s , t) ( 2 ≤ s ≤ t and s=1,t=2) has this property under the comparison model. [ABSTRACT FROM AUTHOR]

Connectivity plays an crucial role in measuring the fault tolerance of a multiprocessor system in the case of vertices failures. Extra connectivity is an important indicator of a network's ability for diagnosis and fault tolerance. In this paper, we analyse the fault tolerance ability of the n , k -dimensional enhanced hypercube, a well-known interconnection network denoted by Q n , k proposed for multiprocessor systems, and establish the { 1 , 2 , 3 } -extra connectivity of Q n , k . [ABSTRACT FROM AUTHOR]

Connectivity is an important measure to explore the fault tolerance and reliability of the network structure based on a graph model. Let G = (V , E) be a connected graph. A r-component cut of G is a set S of vertices, G−S has at least r components. The r-component connectivity c κ r (G) of G is the size of the smallest r-component cut. The r-component edge connectivity c λ r (G) can be defined similarly. In this paper, we determine the r-component connectivity of Cayley graphs generated by transposition trees c κ r + 1 (T n) = r (n − 1) − r (r + 1) / 2 + 1 for r = 1 , 2 , ... , n − 1 and n ≥ 3 ; we also obtain the r-component edge connectivity of Cayley graphs generated by transposition trees T n for small r and the upper and lower bounds of c λ r + 1 (T n) for r = 1 , 2 , ... , n − 1 and n ≥ 3. [ABSTRACT FROM AUTHOR]

Diagnosability is an important parameter to measure the fault tolerance of interconnection networks. Arrangement graph A n , k is a generalisation of the star graphs, yet it is more flexible in its size than the star graphs. In this paper, we study the local diagnosability of A n , k , and show that it has the strong local diagnosability property even if there exist (k − 1) (n − k) − 1 missing edges in it under the MM* model, and the result is optimal with respect to the number of missing edges. [ABSTRACT FROM AUTHOR]

Over the past few years, capturing the characteristics of the Gnutella overlay network has been one of the major research activities in distributed computing. While most of the works are focusing on the node degree distribution, upstream/downstream traffics and file distribution, only a few have been done on modeling the session lifetime distribution. A good model of the session lifetime distribution is basically a key to understand the formation and evolution of the topological structure of Gnutella. In this paper, the main objective is to present a measurement result we did in 2003 on the session lifetime distribution and make comparison with the results obtained by other researchers during 2001-2005. In accordance with our measurement, it is found that the lifetime distribution fits to a power law distribution with exponential cut-off. Specifically, for all t ≥ 15 min, [image omitted] . By inspecting the slopes of the log-log curve (i.e. [image omitted] ) at 20 min (i.e. short lifetime) and 600 min (i.e. long lifetime), it is also found that the shape of our model matches to those curves obtained by other researchers based on the measurements in 2001, 2002 and 2004. Therefore, it is observed that the session lifetime distribution of Gnutella might be an invariant characteristic independent of the technological advancement or protocol change during the period 2002-2005. [ABSTRACT FROM AUTHOR]

Clustering techniques are very attractive for identifying and extracting patterns of interests from datasets. However, their application to very large spatial datasets presents numerous challenges such as high-dimensionality, heterogeneity, and high complexity of some algorithms. Distributed clustering techniques constitute a very good alternative to the Big Data challenges (e.g., Volume, Variety, Veracity, and Velocity). In this paper, we developed and implemented a Dynamic Parallel and Distributed clustering (DPDC) approach that can analyse Big Data within a reasonable response time and produce accurate results, by using existing and current computing and storage infrastructure, such as cloud computing. The DPDC approach consists of two phases. The first phase is fully parallel and it generates local clusters and the second phase aggregates the local results to obtain global clusters. The aggregation phase is designed in such a way that the final clusters are compact and accurate while the overall process is efficient in time and memory allocation. DPDC was thoroughly tested and compared to well-known clustering algorithms BIRCH and CURE. The results show that the approach not only produces high-quality results but also scales up very well by taking advantage of the Hadoop MapReduce paradigm or any distributed system. [ABSTRACT FROM AUTHOR]

In this survey paper, we compare native double precision solvers with emulated- and mixed-precision solvers of linear systems of equations as they typically arise in finite element discretisations. The emulation utilises two single float numbers to achieve higher precision, while the mixed precision iterative refinement computes residuals and updates the solution vector in double precision but solves the residual systems in single precision. Both techniques have been known since the 1960s, but little attention has been devoted to their performance aspects. Motivated by changing paradigms in processor technology and the emergence of highly-parallel devices with outstanding single float performance, we adapt the emulation and mixed precision techniques to coupled hardware configurations, where the parallel devices serve as scientific co-processors. The performance advantages are examined with respect to speedups over a native double precision implementation (time aspect) and reduced area requirements for a chip (space aspect). The paper begins with an overview of the theoretical background, algorithmic approaches and suitable hardware architectures. We then employ several conjugate gradient (CG) and multigrid solvers and study their behaviour for different parameter settings of the iterative refinement technique. Concrete speedup factors are evaluated on the coupled hardware configuration of a general-purpose CPU and a graphics processor. The dual performance aspect of potential area savings is assessed on a field programmable gate array (FPGA). In the last part, we test the applicability of the proposed mixed precision schemes with ill-conditioned matrices. We conclude that the mixed precision approach works very well with the parallel co-processors gaining speedup factors of four to five, and area savings of three to four, while maintaining the same accuracy as a reference solver executing everything in double precision. [ABSTRACT FROM AUTHOR]

The optical bus based arrays have recently shown many applications. In this paper we use the pipelining ability of the LARPBS to solve the repetitions detection problem. We propose an algorithm that finds all kind of repetitions in a string including k-power repetitions, overlapping repetitions and non-contiguous repetitions. For a given distance, this operation is performed in O(1) communication cycle with a constant computation time processors on a O(n) processors LARPBS for a string of length n. For all the periods, it is O(n) bus cycle. No previous algorithms were known for this problem on the LARPBS. [ABSTRACT FROM AUTHOR]

In this paper, we present an emergent computing architecture scheme, denoted as marching pixels (MPs), which is well-suited for massively-parallel embedded systems based realising smart optical sensors. Smart in this context refers to sensors in which signal capturing and signal processing is combined either in one chip or in a compact 3D chip stack. MPs are life-like agents, which are crawling within a pixel field in such a chip in order to find centre points and other attributes of objects in binary images. MPs fulfill these tasks only by utilising local interaction and stigmertic communication. Principal ideas, the architecture and first results are described in this paper along with a comparison to both conventional and other unconventional computing approaches. [ABSTRACT FROM AUTHOR]

In this paper, we investigate algorithms for generating communication code to run on distributed-memory systems. We modify algorithms from previously published work and prove that the algorithms produce correct code. We then extend these algorithms to incorporate the mapping of virtual to physical processors and prove the correctness of this extension. This technique can reduce the number of interprocessor messages. In the examples that we show, the total number of messages was reduced from O( N 2 ) to O( P 2 ), where N is the input size and P is the number of physical processors. The reason that it is important to revisit communication code generation and to introduce a formal specification of the incorporation of mapping in the communication code generation is so that we can make use of the many scheduling heuristics proposed in the literature. We need a generalized mapping function so that we can apply different mapping and scheduling heuristics proposed in the literature for each input program, therefore improving the average performance. [ABSTRACT FROM AUTHOR]

The nonuniversality in computation theorem (NCT) states that no computer capable of a finite and fixed number of basic operations per time unit can be universal. This result, obtained in 2005, disproves major prevailing dogmas in computer science, on the strength of several counterexamples that differ significantly from one another. Thus, the NCT shows that the Church–Turing thesis (CTT) is false: It is not true that the Turing Machine can execute any computation possible on any other computer. It also disproves the inflated CTT, whereby there exists a universal computer, that is, a physical device capable of carrying out any computation conceivable. At the heart of the NCT is a refutation of the simulation principle, which states that any computation possible on a general-purpose computercan be simulated, more or less efficiently, on any other general-purpose computer. While more than 10 years have now elapsed since the NCT was established, the result is still widely misunderstood. This state of affairs is due to a number of misconceptions about the nature of the counterexamples and the significance of the NCT. The purpose of this paper is to dispel these misconceptions and convey the simple, yet important idea that universality in computation cannot be achieved, except by a computer capable, each time unit, of an infinite number of basic operations executed in parallel. [ABSTRACT FROM PUBLISHER]

One of the most important issues in the design of distributed systems is process scheduling, which maps applications to resources in an attempt to reduce the application execution time or maximise resource utilisation. The complexity involved in finding good scheduling solutions has motivated the design of several heuristics and approximation algorithms. Although heuristics aim at finding good solutions within acceptable time constraints, they do not guarantee solution quality. On the other hand, approximation-based algorithms provide optimal solution bounds, but they are usually designed to address simple scenarios. In an attempt to combine the best of both algorithmic approaches, this paper proposes the min-heap-based scheduling algorithm (MHSA), which finds solutions for homogeneous and heterogeneous distributed environments within acceptable time constraints and optimal solution bounds. MHSA considers applications that can be bag-of-tasks or which communicate among each other. Such solutions minimise the application execution time and maximise resource utilisation; therefore, MHSA is a system-oriented scheduling algorithm. Results confirm that MHSA provides better solutions than schedulers such as Random, Round-Robin, Workload-Queue and List-Scheduling. In addition, MHSA provides an-approximation ratio for reaching optimal solution bounds. [ABSTRACT FROM AUTHOR]

The current Internet has several known challenges, such as routing scalability, mobility, multihoming, traffic engineering, etc. due to the overloaded semantics of IP address, i.e. it is used as a node identifier (ID) and a node locator (LOC). Thus, the research community has redesigned the Internet architecture based on ID/LOC separation to overcome the limitations of the current Internet. In all Internet architectures based on ID/LOC separation, ID to LOC mapping system is necessarily required to bind ID and its LOC, since ID is no longer dependent to its LOC logically or physically. Thus, how to design the mapping system is a key challenge in ID/LOC separation architecture. In this paper, we analyse qualitatively the mapping systems proposed in ID/LOC separation architectures to provide insights into designing a new mapping system. The main contribution in this paper is that we categorise ID to LOC mapping systems according to the mapping server structure and provide the pros and cons of the mapping systems belonging to each category. Based on our qualitative analysis, we also examine intuitively if the mapping systems in each category satisfy their requirements. [ABSTRACT FROM AUTHOR]

Quadratic Unconstrained Binary Optimization (QUBO) is a combinatorial optimization problem defined by an energy function that consists of a quadratic formula involving multiple binary variables. We propose the method of combing bit reduction and duplication technique that can generate hard instances of QUBO problems from any original QUBO problem without changing the size. The idea is to reduce the original QUBO problem for a specified number of bits and duplicate the same number of bits while ensuring all duplicated pair of bits take the same binary values. Experimental results show that generated QUBO instances are much hard for solving. [ABSTRACT FROM AUTHOR]

We propose a novel blockchain-based academic degree verification system. The system relies on a smart contract deployed on the Ethereum blockchain platform. The system protects the privacy of the students who have been verified without the need of encryption, and hence, avoid the complexity of security key management and the potential disaster if the keys are stolen or revealed. The amount of degree data is also designed to be minimum for proper verification. All operations on the critical path of the system are automated without any human intervention. However, the system does assume that the organization would stand behind the proposed system to ensure the system's integrity. The system is open to anyone who wishes to verify the degree of a student. To prevent spamming and to support the long-term sustainability of the system, a small fee is charged for each verification request. A portion of the fees collected is immediately credited to the participating schools. To facilitate automated school enrollment into the system, any school may request to enroll by placing a significant amount of stake in ETH. The stake can be refunded when the school decides to leave the system. However, if the school is blacklisted due to misbehavior, the stake is confiscated. This design strongly discourages adversaries from abusing the system. The performance of the system is fully characterized with regard to the gas consumption of key operations in various use cases. We demonstrate that the system is conducive for long-term sustainability. [ABSTRACT FROM AUTHOR]

We present a paradigm for access control to typed objects logically connected to form a hierarchy, whereby each object has a single parent, and may have children. Protection domains are divided into three classes. A vertical domain includes authorizations for the objects in a subhierarchy. A horizontal domain includes authorizations for the children of a given object. A simple domain includes an authorization for a single object. We introduce the hierarchical capability concept as a generalization of the classical password capability concept. A hierarchical capability references an entire domain instead of a single object. [ABSTRACT FROM AUTHOR]

For scientific software, especially those used for large-scale simulations, achieving good performance and efficiently using the available hardware resources is essential. It is important to regularly perform benchmarks to ensure the efficient use of hardware and software when systems are changing and the software evolves. However, this can become quickly very tedious when many options for parameters, solvers, and hardware architectures are available. We present a continuous benchmarking strategy that automates benchmarking new code changes on high-performance computing clusters. This makes it possible to track how each code change affects the performance and how it evolves. [ABSTRACT FROM AUTHOR]

This article develops a new control-oriented linear parameter varying (LPV) model, which is a linear state-space model with varying parameters aimed at efficiently representing large-scale offshore wind turbines (LOWTs). Firstly, the LPV model is constructed by the mechanism analysis of a 15-MW direct-drive permanent magnet wind turbine. Subseuently, based on the operational data from FAST, the parameters with unknown values in the LPV model are determined by gain scheduling and parameter estimation. Furthermore, a deloading operation for the active output power regulation of LOWTs is proposed based on coordinating the rotor over-speed control (ROC) and the pitch angle control (PAC). [ABSTRACT FROM AUTHOR]

Public and permissionless blockchain systems are challenged by Sybil attacks, in which attackers use multiple identities to gain control. Traditionally, such attacks are prevented by consensus mechanisms relying on resource expenditure. However, such mechanisms (e.g. proof of work) face criticism for being wasteful. To address this and other concerns, novel blockchain systems backed by new consensus mechanisms have recently emerged. We formalise three key characteristics pursued by these systems: permissionlessness, Sybil attack resistance, and freeness. We demonstrate that no blockchain protocol can simultaneously achieve all three characteristics within the paradigm established by our formalisation. Thus, a trilemma emerges for distributed ledger technology designers, who must balance these characteristics thoughtfully. [ABSTRACT FROM AUTHOR]

Consider the empirical laws established in the History of Science. The process of scientific discovery can be roughly explained by (Popper-Kuhn's) cycle which starts with collecting evidence, introducing a model to explain observations, predicting further observations, and confronting them with experimentation, reinforcing or disproving the model. We attempt to prove that the empirical evidence modeled by arithmetical operations can be automatically discovered with a small amount of information, i.e. with very short Blums' locking sequences, due to the suitable enumerations of arithmetical functions. [ABSTRACT FROM AUTHOR]

Mobile ad hoc networks (MANET) are becoming increasingly popular, particularly in the realm of sensor networks. However, most of the ongoing research is centered on protocol development or data delivery characteristics, particularly in the areas of throughput and latency. But in the realm of wireless networks, a new metric needs to be considered, one that takes into consideration the power efficiency across the wireless network. Wireless nodes have a finite amount of energy to operate with, and when that energy is depleted, the node dies and no longer performs its intended functions. In order to maintain a functioning MANET for longer periods of time, the network as a whole needs to be optimized for power efficiency. Ongoing research in the realm of MANETs includes the development of power-aware routing algorithms. A number of them have been developed, often augmenting existing routing protocols. To aid in the validation of network performance improvement, this paper describes a metric that incorporates both power efficiency and data throughput of MANETs. This new network performance metric can be used for selecting the optimal routing protocol for the network scenario. [ABSTRACT FROM AUTHOR]

Metaheuristics have been showing interesting results in solving hard optimization problems. However, they become limited in terms of effectiveness and runtime for high dimensional problems. Thanks to the independency of metaheuristics components, parallel computing appears as an attractive choice to reduce the execution time and to improve solution quality. By exploiting the increasing performance and programability of graphics processing units (GPUs) to this aim, GPU-based parallel metaheuristics have been implemented using different designs. Recent results in this area show that GPUs tend to be effective co-processors for leveraging complex optimization problems. In this survey, mechanisms involved in GPU programming for implementing parallel metaheuristics are presented and discussed through a study of relevant research papers. Metaheuristics can obtain satisfying results when solving optimization problems in a reasonable time. However, they suffer from the lack of scalability. Metaheuristics become limited ahead complex high-dimensional optimization problems. To overcome this limitation, GPU based parallel computing appears as a strong alternative. Thanks to GPUs, parallel metaheuristics achieved better results in terms of computation, and even solution quality. [ABSTRACT FROM AUTHOR]

In Optical WDM networks, a single fiber carries enormous amount of data and consequently designing schemes to prevent the user traffic from disruption and techniques to recover from failures still remains an important area of research. In this paper we address the problem of protecting all-optical WDM mesh networks from single link failure. We propose an offline protection scheme that is not only cost-efficient in terms of network resource consumption but can also provide quick recovery from a link failure. In our scheme, backup paths are combined to reduce cost of protection and to facilitate fast recovery from a failure. We address a critical issue related to selection of backup paths to be combined, which was not addressed earlier. Instead of choosing backup paths for connection requests arbitrary as done in an existing scheme, our scheme is designed to select backup paths in a way that they are suitable for combination and can maximize resource saving as well. We first provide an Integer Linear Program for the problem, and then propose a heuristic eStreams that can provide near optimal solution in polynomial time. Performance comparisons with some well-known schemes show that our protection scheme is definitely an improvement in terms resource conservation while ensuring fast recovery from a link failure. The figure shows how a primary (working) path is protected with the help of a link-disjoint backup path in WDM optical networks. A primary path is used to carry traffic under normal fault free condition. In the event of a link fault (fiber-cut) the traffic is switched over to the backup path. [ABSTRACT FROM AUTHOR]

Cloud computing providers have disseminated dynamic storage provisioning delivered to end users as on-demand services. Although cloud file storage and sharing has become popular among home users, the access requirements, performance expectations and usage characteristics are different for organisations, and were not originally considered by popular applications and tools for synchronising files between cloud providers and local repositories. Moreover, multisite organisations traditionally have legacy file storage and wide-area networking solutions to support their business systems. Typically, the file repositories are replicated between sites using private communication links. The combination of legacy storage solutions interconnected through private links with cloud-based file storage is a challenging task. In this context, this paper introduces Cloud4NetOrg, a client architecture for cloud file storage and multisite repository synchronisation. We implemented prototypes of this architecture that interact with two popular cloud file services (DropBox and OneDrive), and the experimental results indicate a promising application in collaborative environments with several LANs. Indeed, Cloud4NetOrg decreases the synchronisation time and the total data transferred from/to cloud repositories by using the organisation repositories as a hierarchical cache system. Cloud4NetOrg is proposed for geographically distributed organisations composed of dynamic and temporary collaborative groups. The interaction between employees is based on file sharing. Commonly, sites are interconnected by a private network and have an internal data storage repository. A single site can have multiple subnetworks to interconnect the collaborative groups. In addition, home-office users collaborate through the Internet, usually using size-limited storage devices. [ABSTRACT FROM AUTHOR]

Sensor deployment positions play an important factor in determining target location estimation error performance in sensor networks employing received signal strength indicator measurements. The problem we investigate is as follows: Given a deployment area that has possibly non-uniform estimation error requirements and some initial sensor positions, then how can these sensors be deployed such that requirements are met as best as possible? We propose three variants of a low-complexity distributed strategy that require little information exchange between neighboring sensors. Each sensor locally calculates both the direction and magnitude of movement necessary to reduce the difference between achieved and required estimation errors at points within it sensing radius. The direction of movement is calculated as a weighted combination of these points. The weights can incorporate the error difference, number of sensors covering a point, the distance to a sensor and sensor density. Depending on the parameters chosen, three different weighing methods (namely, the density weighted centroid, error weighted centroid (EWC) and modified EWC) are proposed in this paper. We also provide an analytic derivation of the necessary distance a sensor should move. The proposed strategy is compared against a the RELOCATE algorithm and centralized generic genetic algorithm (GA) relocation method. Simulation results demonstrate that using the distributed strategy can achieve a comparable performance (within 10%) of the GA's performance and significantly outperform that of the RELOCATE. Sensor movement for different k. [ABSTRACT FROM AUTHOR]

The idea of employing scavenged energy from human motion to run electronic devices is attracting increased attention of many researchers worldwide. However, there is still limited knowledge of energy characteristics generated by human motions. Moreover, level of human activities varies during a day from sitting for several hours to running on a treadmill. This highlights a vital need for energy harvesting modelling. Hence, in this paper, we aim to investigate the feasibility of running a simple fitness gadget with scavenged energy from human motions. We analyse kinetic energy generated by human activities and develop energy harvesting modelling techniques which estimate the required time for a device to store needed energy between each active periods. Two large, real-world data sets are employed to obtain empirical results for different motion scenarios. We believe these results are informative to the design of energy harvesting fitness gadgets with sensing applications. We consider two large data sets consist of acceleration signals of walking and human daily routines. Acceleration signals are converted to power in order to estimate potential energy that can be generated from human motions. We also propose energy harvesting modelling techniques for self-powered fitness gadgets. By considering potential application for fitness gears, these techniques are evaluated using statistical analysis and different motion scenarios. [ABSTRACT FROM AUTHOR]

We propose a new single-electron circuit (SEC) that mimics the behaviours of slime molds. The SEC has unique properties and many advantages including extremely low power consumption. However, the most appropriate information processing way on the SEC has not been established yet. As one approach to find the suitable way, we focus on a natural phenomenon, specifically, the behaviours of slime molds. These molds can feed efficiently and their behaviours can be regarded as performing information processing. Their information-processing-like behaviours can be divided into three functions. Therefore, if we can express these functions on the circuit, it can mimic the slime mold behaviour, i.e. it can operate as a slime-mold-inspired information-processing circuit. In this study, we designed a slime-mold-inspired SEC that utilised three functions based on three types of behaviours, and attempted to imitate these functions by using a multi-layered structure. Our circuit has four layers; three for mimicking the divided functions and the last for output. We evaluated its operation by Monte Carlo simulation and confirmed its operation was correct. This paper indicates design of a new single-electron circuit (SEC). For the SEC to process information efficiently, we focus on the behaviors of slime molds. It is known that these molds can feed efficiently and their behaviors can be regarded as performing information processing. In this study, we designed a slime-mold-inspired SEC having a multi-layered structure. We evaluated its operation by Monte Carlo simulation and confirmed its operation was correct. Figure shows a sample operation of our circuit. (a) is the input pattern and (b) is the output pattern, respectively. The optimal path was revealed. [ABSTRACT FROM AUTHOR]

In a swarm system, for example in a beehive, group decision is based on interactions and interferences of all individuals without a central unit that decides for everybody. When making experiments with young honeybees (Apis mellifera), a swarm algorithm, called BEECLUST, was derived. The algorithm enables swarms to locate the 'Global-Goal' out of several local optima. There were also four different behavioural types discovered during the experiments: Random-Walker, Goal-Finder, Wall-Follower and the Immobile Bee. In this paper, we introduce the four behavioural types to the BEECLUST algorithm and analyse how the decision making process of the swarm can be influenced. We show how the different types can be used to optimise the decision making for a certain setup of the arena and discuss about Swarm Level Optimisation. [ABSTRACT FROM AUTHOR]

Named data networking (NDN) is a recently proposed networking paradigm that aims to overcome the limitations of the Internet Protocol, while preserving its strengths by using named data rather than named hosts for communication. In NDN, the content is determined by its name, and each NDN packet has a name that determines the content it is bringing or forwarding. In this new paradigm, content names have a hierarchical structure and variable length and, therefore, these features will introduce new challenges to satisfy reasonable speed at large scale. In this paper we propose, design and implement a new approach called fast two-dimensional filter with hash table (FTDF-HT). In the proposed approach, a hash table storing the name prefixes, in a hierarchical manner, is only accessed when the proposed filter (FTDF) states that the name under querying exists. Thus, reducing the unnecessary access to the hash table. Moreover, the access to the hash table will be done with same hash function that is used for FTDF and this will minimize the search time. Our evaluation results show that FTDF-HT can achieve high lookup speed and exhibit good scalability to large-scale prefixes table. [ABSTRACT FROM AUTHOR]

The concept of smart grid (SG) has been introduced to improve the operation of the power systems. In modern structures of power systems, different reasons prompt researchers to suggest integrated analysis of multi-carrier energy systems. Considering synergy effects of the couplings between different energy carriers and utilising intelligent technologies for monitoring and controlling of energy flow may change energy system management in the future. In this paper, we propose a new solution which is entitled ‘smart energy hub’ (SEH) that models a multi-carrier energy system in a SG. SEH solutions allow homeowners to manage their energy consumption to reduce their electricity and gas bill. We present this concept for a residential customer by an ‘energy management system’ which uses reinforcement learning algorithm and Monte Carlo estimation method for finding a near optimal solution. The simulation results show that by using this concept and then applying the algorithm for a residential customer, running costs are reduced up to 40% while keeping the household owner's desired comfort levels. [ABSTRACT FROM PUBLISHER]

The accurate forecasting of solar irradiance with hybrid machine learning algorithm is presented in this paper. A novel Persistence-Extreme Learning Machine (P-ELM) algorithm is used for training of the system. The Clearness Index (CI) value of the 22 districts of Andhra Pradesh (India) is calculated and out of which four areas are identified with highest CI values. The Global Horizontal Irradiance (GHI) is forecasted for the selected areas with different weather conditions such as winter, summer and rainfall seasons using a P-ELM algorithm. The input parameters are Temperature, Diffuse horizontal irradiance, pressure and past GHI and GHI for the next instant as the output is considered. The real time data is obtained for every one hour interval for a period of one month. The performance of the P-ELM algorithm is evaluated in terms of Mean Absolute Error and Root Mean Square Error. From the obtained results, it is observed that P-ELM algorithm offers better performance over the fundamental P-ELMs. The P-ELM algorithm gives good forecasting accuracy with minimum simulation time. The simulation of P-ELM algorithm is carried out using MATLAB 2013a environment. The P-ELM algorithm is very much beneficial for accurate and reliable real time solar forecasting. [ABSTRACT FROM AUTHOR]

In this paper, we study the problem of decentralized learning in sensor networks in which local learners estimate and reach consensus to the quantity of interest inferred globally while communicating only with their immediate neighbours. The main challenge lies in reducing the communication cost in the network, which involves inter-node synchronisation and data exchange. To address this issue, a novel asynchronous broadcast-based decentralized learning algorithm is proposed. Furthermore, we prove that the iterates generated by the developed decentralized method converge to a consensual optimal solution (model). Numerical results demonstrate that it is a promising approach for decentralized learning in sensor networks. [ABSTRACT FROM AUTHOR]

Most of the metaheuristics can efficiently solve unconstrained problems; however, their performance may degenerate if the constraints are involved. This paper proposes two constraint handling approaches for an emerging metaheuristic of Cohort Intelligence (CI). More specifically CI with static penalty function approach (SCI) and CI with dynamic penalty function approach (DCI) are proposed. The approaches have been tested by solving several constrained test problems. The performance of the SCI and DCI have been compared with algorithms like GA, PSO, ABC, d-Ds. In addition, as well as three real world problems from mechanical engineering domain with improved solutions. The results were satisfactory and validated the applicability of CI methodology for solving real world problems. [ABSTRACT FROM AUTHOR]

Increasing system complexity, energy and device reliability, requirement of modular approach, structured layout, effective spatial reuse of resources, scalability and re-programmability have made network-on-chip (NoC) an obvious interconnection design alternative to the ubiquitous bus based on chip communication architecture in system-on-chip. Designing of a topology and its routing scheme plays a vital role in determining performance of any NoC architecture. In recent years, 3D stacked NoC architecture attracts added interest in NoC design as it offers improved performance and shorter global interconnect. In this paper, we have developed a partially, vertically interconnected 3D topology, namely 3D Recursive Network Topology (3D RNT) and prove that the topology has a Hamiltonian connectedness. We have developed deadlock-free routing algorithm for the 3D RNT topology. Also, we compare the performance of the 3D RNT with partially and fully connected 3D mesh topologies (3D PMT and 3D FMT) by conducting suitable experiments. The experiment results show that there is not much deviation in respect of the performance of the 3D RNT on comparing with 3D PMT and 3D FMT even though a number of vertical links are trimmed down to 75%, which is an encouraging outcome as far as design space is concerned. [ABSTRACT FROM AUTHOR]

The main contribution of this paper is to introduce two parallel memory machines, the discrete memory machine (DMM) and the unified memory machine (UMM). Unlike well-studied theoretical parallel computational models such as parallel random access machines, these parallel memory machines are practical and capture the essential feature of the memory access by graphical processing units (GPUs). As a first step of the development of algorithmic techniques on the DMM and the UMM, we first evaluate the computing time for the contiguous access and the stride access to the memory on these models. We then present parallel algorithms to transpose a 2D array on these models and evaluate their performance. Finally, we show that, for any permutation given in offline, data in an array can be moved efficiently along the given permutation both on the DMM and on the UMM. Since the computing time of our permutation algorithms on the DMM and the UMM is equal to the sum of the lower bounds obtained from the memory bandwidth limitation and the latency limitation, they are optimal from the theoretical point of view. We believe that the DMM and the UMM can be good theoretical platforms to develop algorithmic techniques for GPUs. [ABSTRACT FROM AUTHOR]

The concept of unreliable failure detector was introduced by Chandra and Toueg as a mechanism that provides information about process failures. This mechanism has been used to solve several agreement problems, such as the consensus problem. In this paper, algorithms that implement failure detectors in partially synchronous systems are presented. First two simple algorithms of the weakest class to solve the consensus problem, namely the Eventually Strong class (⋄S), are presented. While the first algorithm is wait-free, the second algorithm isf-resilient, wherefis a known upper bound on the number of faulty processes. Both algorithms guarantee that, eventually, all the correct processes agree permanently on a common correct process, i.e. they also implement a failure detector of the class Omega (Ω). They are also shown to be optimal in terms of the number of communication links used forever. Additionally, a wait-free algorithm that implements a failure detector of the Eventually Perfect class (⋄P) is presented. This algorithm is shown to be optimal in terms of the number of bidirectional links used forever. [ABSTRACT FROM AUTHOR]

The author discusses articles published in the journal including one on new web segmentation methodology based on the image processing technology , one on field programmable gate array (FPGA) and one on high performance storage system for Resource Description Format (RDF) documents.

Network-on-Chip provides a packet-based and scalable inter-connected structure for spiking neural networks. However, existing neural mapping methods just distribute all neurons of a population into an on-chip network core or nearby cores sequentially. As there is no connection among population, the population based mapping degrades inter-neuron communicating performance between different cores. This paper presents a Cross-LAyer based neural MaPping method that maps synaptic connected neurons belonging to adjacent layers into the same on-chip network node. In order to adapt to various input patterns, the strategy also takes input spike rate into consideration and remap neurons for improving mapping efficiency. The method helps to reduce inter-core communication cost. The experimental results demonstrate the efficient results of the proposed mapping strategy in the aspect of spike transfer latency as well as dynamic energy cost improvement. In the applications of handwritten digits and edge extraction, in which the type of interconnection among neurons is different, the neural mapping algorithm reduces spike average transfer latency by maximum 42.83%, and reduces dynamic energy by maximum 36.29%. [ABSTRACT FROM AUTHOR]