Search

As the number of Internet services is growing rapidly, the network users need a framework to locate and utilize those services that are available in the Internet. As a result, dynamic service discovery plays an important role in certain networks, such as P2P networks. Recently many P2P mechanisms have been proposed which focus on the naming and discovery protocols. However, because these mechanisms do not provide congestion management and priority-based scheduling in P2P networks, they cannot provide an efficient communication among the nodes. Based on these observations, in this paper, we propose a QoS-based agent framework, designed to resolve such problems, and demonstrate its effectiveness. [ABSTRACT FROM AUTHOR]

This paper demonstrates a novel car-following model based on driver or passengers' comfort. As we know, hasty deceleration during emergency brake will cause passengers feel uncomfortable. According to the relationship between brake acceleration and people's comfortable feeling, the comfortable model is setup. The model calculates the following car's acceleration by measuring the distance between the following car and the preceding car, the velocity of the following car, and controls the car's acceleration to make driver and passengers feel comfort. The paper combine the model with the pervasive computing concept, provoke the pervasive computing driver behavior modeling idea and turn it into reality to increase the adaptability and reliability of car's parts, when car equipped with this device, the prospect is not only the assistant driver or comfortable driver are realized in the car-following circumstance, but also the whole car's performance will be improved. [ABSTRACT FROM AUTHOR]

This paper suggests a context adaptive self-configuration system, in which the system itself is aware of its available resources, and executes a configuration suitable for each system and the characteristics of each user. A great deal of time and effort is required to manually manage computing systems, which are becoming increasingly larger and complicated. Many studies have dealt with these problems but most unfortunately have focused on ‘automation', which do not reflect the system specifications that differ between systems. Therefore, this paper proposes an adaptive self-configuration system that collects diverse context information for a user and its system, reflects it to an auto-response file and executes the detail parameter setting automatically. A prototype was developed to evaluate the system and compare the results with the conventional methods of manual configuration and MS-IBM systems. The results confirmed the effectiveness of the proposed system. [ABSTRACT FROM AUTHOR]

This paper presents an abstract model for service compositions based on agents. The model is based on distributed Abstract State Machine and tallies with Universal Plug and Play (UPnP) architecture standard. In the abstract model, the composition of services can be viewed as a union of agents. This model can be used to depict a number of Web Service applications, and as an instance, the paper presents a BPEL Abstract Machine. [ABSTRACT FROM AUTHOR]

Sequential pattern mining is an important data mining problem with broad applications. Especially, it is also an interesting problem in virtual environments. In this paper, we propose a projection-based, sequential patterngrowth approach, called PrefixUnion. Meanwhile, we also introduce the relationships among transactions, views and objects. According to these relationships, we suggest two mining criteria — inter-pattern growth and intra-pattern growth, which utilize these characteristics to offer ordered growth and reduced projected database. As a result, the large-scale VRML models could be accessed more efficiently, allowing for a real-time walk-through in the scene. [ABSTRACT FROM AUTHOR]

The importance of problem solving environment (PSE)s over Grid has been emphasized owing to the heterogeneity and the large volume of resources involved and the complexity of computational applications. This paper proposes a Grid-enabled PSE called as Computing Environment for Grid Applications (CEGA) and discusses how it is evolving to develop a computational application in a style of workflow model and incorporates Grid computing services to extend its range of services and handle information for development, deployment, execution and maintenance for an application as well as an application requirement itself. The paper describes the architecture of CEGA and its implementation for development, execution and visualization of an application. [ABSTRACT FROM AUTHOR]

This paper shows the mathematical development to derive the integral-differential equations and the algorithms implemented in MATLAB to predict the cutting force in real time in high speed machining processes. This paper presents a cutting-force-based model able to describe a high-speed machining process. The model considers the cutting force as an essential output variable in the physical processes taking place in high-speed machining. For the sake of simplicity, only one type of end mill shapes is considered (i.e., cylindrical mill) for real-time implementation of the developed algorithms. The developed model is validated in slot-milling operations. The results corroborate the importance of the cutting-force variable for predicting tool wear in high-speed machining operations. Keywords: modeling; complex systems; high-speed machining. [ABSTRACT FROM AUTHOR]

This paper presents two very efficient sorting algorithms. MSL is an O(N*B) in-place radix sorting algorithm, where N is the input size and B is the keys length in bits. This paper presents an implementation of MSL that is sub-linear in practice, for uniform data, on Pentium 4 machines. We also present an O(N*logN) hybrid quicksort that has a non-quadratic worst case. [ABSTRACT FROM AUTHOR]

In this paper, we present an analysis on optimization and risk management in Communication Networks (CNs). The model is proposed for offline traffic engineering optimization, which takes a centralized view of bandwidth allocation, performance control, and risk control of network profit shortfall. First, we introduce a linear penalty cost in the CN optimization model and derive the optimal bandwidth capacity with the penalty cost. Then, we use the mean-variance approach to analyze the profit shortfall risk in CNs. Finally, numerical results are shown to reveal the impact of the penalty cost on the CNs performance. [ABSTRACT FROM AUTHOR]

This paper presents a Framework for Distribute Transaction processing over Grid Environment, called DCP-Grid. DCP-Grid complements Web Services with some OGSI functionalities to implement the Two Phase Commit (2-PC) protocol to manage two types of Distribute Transactions, Concurrent and Conversational transactions, properly in this kind of environment. Although DCP-Grid is still under development at the Universidad Politécnica de Madrid, in this paper, we present the design and the general characteristics associated to the implementation of our proposed Framework. [ABSTRACT FROM AUTHOR]

With the current popularity of cluster computing systems, it is increasingly important to understand the capabilities and potential performance of various interconnection networks. In this paper, we propose an analytical model for studying the capabilities and potential performance of interconnection networks for multi-cluster systems. The model takes into account stochastic quantities as well as network heterogeneity in bandwidth and latency in each cluster. Also, blocking and non-blocking network architecture model is proposed and are used in performance analysis of the system. The model is validated by constructing a set of simulators to simulate different types of clusters, and by comparing the modeled results with the simulated ones. [ABSTRACT FROM AUTHOR]

Data Mining is often required to be performed among a number of groups of sites, where the precondition is that no privacy of any site should be leaked out to other sites. In this paper, a hierarchical infrastructure is proposed for large-scale distributed Privacy Preserving Data Mining (PPDM) utilizing a synergy between P2P and Grid. The proposed architecture is characterized with (1) its ability for preserving the privacy in data mining; (2) its ability for decentralized control; (3) its dynamic and scalable ability; (4) its global asynchrony and local communication ability. An algorithm is described to show how to process large-scale distributed PPDM based on the infrastructure. The remarks in the end show the effectiveness and advantages of the proposed infrastructure for large-scale distributed PPDM. [ABSTRACT FROM AUTHOR]

In this paper, a support vector machine (SVM) with polynomial kernel function based nonlinear model multi-step-ahead controller is presented. A SVM based multi-step-ahead predictive model is established by black-box identification. And control law is obtained by numerical and optimization methods respectively. The effect of controller is demonstrated on a recognized benchmark problem. Simulation results show that multi-step-ahead predictive controller can be well applied to nonlinear system with good performance. [ABSTRACT FROM AUTHOR]

RNA structure prediction remains one of the most compelling, yet elusive areas of computational biology. Many computational methods have been proposed in an attempt to predict RNA secondary structures. A popular dynamic programming (DP) algorithm uses a stochastic context-free grammar to model RNA secondary structures, its time complexity is O(N4) and spatial complexity is O(N3). In this paper, a parallel algorithm, which is time-wise and space-wise optimal with respect to the usual sequential DP algorithm, can be implemented using O(N4/P) time and O(N3/P) space in cluster. High efficient utilization of processors and good load balancing are achieved through dynamic mapping of DP matrix to processors. As experiments shown, dynamic mapping algorithm has good speedup. [ABSTRACT FROM AUTHOR]

This paper presents a novel method for estimating the coordinates of a 3D object using the four vertices of a quadrangle and the camera motion parameters. Estimation of 3D object coordinates from 2D images of video is a studied problem in augmented reality. However, most solutions are dependent on fiducial markers in video or known coordinate systems which are required with superimposition of virtual object on frames. In this paper, we begin with the fact that the rectangular objects in 3D real world are projected the perspective quadrangle onto image planes. We can estimate 3D object coordinates from 4 vertices of quadrangular objects through transformation of image coordinates. The camera motion parameters between pairs of successive frames in a sequence are calculated using epipolar geometry. [ABSTRACT FROM AUTHOR]

In credit card portfolio management, predicting the cardholders' behavior is a key to reduce the charge off risk of credit card issuers. The most commonly used methods in predicting credit card defaulters are credit scoring models. Most of these credit scoring models use supervised classification methods. Although these methods have made considerable progress in bankruptcy prediction, they are unsuitable for data records without predefined class labels. Therefore, it is worthwhile to investigate the applicability of unsupervised learning methods in credit card accounts classification. The objectives of this paper are: (1) to explore an unsupervised learning method: cluster analysis, for credit card accounts classification, (2) to improve clustering classification results using ensemble and supervised learning methods. In particular, a general purpose clustering toolkit, CLUTO, from university of Minnesota, was used to classify a real-life credit card dataset and two supervised classification methods, decision tree and multiple-criteria linear programming (MCLP), were used to improve the clustering results. The classification results indicate that clustering can be used to either as a stand-alone classification method or as a preprocess step for supervised classification methods. Keywords: Credit Card Accounts Classification, Clustering analysis, CLUTO, unsupervised learning method. [ABSTRACT FROM AUTHOR]

The more web data sources provide XML data, the greater information flood problem has been caused. Hence, there have been increasing demands for efficient methods of discovering desirable patterns from a large collection of XML data. In this paper, we propose a new and scalable algorithm, EFoX, to mine frequently occurring tree patterns from a set of labeled trees. The main contribution made by our algorithm is that there is no need to perform any tree join operation to generate candidate sets. [ABSTRACT FROM AUTHOR]

As a featured function of search engine, clustering display of search results has been proved an efficient way to organize the web resource. However, for a given query, clustering results reached by any user are totally identical. In this paper, we explored a user-friendly clustering scheme that automatically learns users' interests and accordingly generates interest-centric clustering. The basis of this personal clustering is a keyword based topic identifier. Trained by users' individual search histories, the identifier provides most of personal topics. Each topic will be the clustering center of the retrieved pages. The scheme proposed distinguishes the functionality of clustering from that of topic identification, which makes the clustering more personal and flexible. To evaluate the proposed scheme, we experimented with sets of synthetic data. The experimental results prove it an effective scheme for search results clustering. [ABSTRACT FROM AUTHOR]

P2P (Peer-to-Peer) applications allow flexible organization, distribution of role to participating peers and the ability to share information with the other peers for group collaborations. As a result, P2P systems are not only gaining importance, but also becoming ubiquitous media for information exchange. JXTA is a P2P application development infrastructure that enables developers to easily create service-oriented software. This paper presents a low-cost, patient-friendly JXTA-based healthcare system, which is comprised of medical sensor modules in conjunction with wireless communication technology. Among the most important activities associated with the healthcare system are the sharing of medical information and collaborative medical work. The JXTA grouping service facilitates to share data in a secure manner under P2P environments. Through JXTA grouping service, we implemented a prototyped system, which improves the ability of the healthcare workers to cope with dynamic situation, which in turn makes it possible to offer more efficient medical services. [ABSTRACT FROM AUTHOR]

In this paper we present p2pCM, a new distributed component-oriented model aimed to structured peer-to-peer grid environments. Our model offers innovative contributions like a lightweight distributed container model, an adaptive component activation mechanism, which takes into account network proximity, and a decentralized component location and deployment service. We believe that all of the features our component-oriented model provides can be very promising for the development of future wide-area distributed applications. [ABSTRACT FROM AUTHOR]

The open and anonymous of grid make the task of controlling access to sharing information more difficult, which cannot be addressed by traditional access control methods. In this paper, we identify access control requirements in such environments and propose a trust based access control framework for grid resource sharing. The framework is an integrated solution involving aspects of trust and recommendation models, based the discretionary access control (DAC), and are applied to grid resource-sharing systems. In this paper, we integrate technology of web services into idea of trust for describing resources. [ABSTRACT FROM AUTHOR]

The Globus Toolkit makes it very easy and comfortable for grid users to develop and deploy grid service. As for the security mechanism, however, only static authentication and coarse-grained authorization mechanism is provided in current Globus Toolkit. In this paper we address the limitations of current security mechanism in the Globus Toolkit and propose a new architecture which provides fine-grained and flexible security mechanism. To implement this without modifying existing components, we make use of the Aspect-Oriented Programming technique. [ABSTRACT FROM AUTHOR]

Despite recent progress in modelling the shape memory alloy (SMA) behaviour, many difficulties remain due to various limitations of the existing free energy models and strong nonlinearity of these nonlinear materials. Phase kinetics of SMA coupled with thermoelastodynamics is still not fully tractable, and one needs to deal with complicated multiscale character of SMA materials requiring a linkage between their microstructure and macroscopic properties. In this paper we develop a new dynamic model of 3D SMA which employs an improved version of the microscopic Landau theory. Essential properties of the single and multivariant martensitic phase transformations are recovered using consistent steps, which eliminates the problem of non-uniqueness of energy partitioning and relaxes the over-sensitivity of the free energy due to many unknown material constants in previously reported models. We exemplify our results on a model for cubic to tetragonal transformations in a rectangular SMA bar by showing key algorithmic steps which can be extended to more complex cases. [ABSTRACT FROM AUTHOR]

Detection of network problems is an important step in automating network management. Early detection of performance degradation can alleviate the last moment hassel of network managers. This paper focuses on a statistical method aiming at detecting changepoint as quickly as possible using Bayes factor along with the binary segmentation procedure modified for fast detection. Computer simulation verifies the effectiveness and correctness of the proposed approach. Keywords: Bayes factor, binary segmentation, changepoint detection, network management, non-homogeneous Possion process. [ABSTRACT FROM AUTHOR]

This paper investigates the static traffic grooming in WDM optical mesh networks. Our objective is to improve the network throughput and to minimize the blocking probability. As take care of this problem efficiently, we propose Shortest-path First Traffic grooming(SFT) algorithm. The comprehensive computer simulation shows that our proposed algorithm is up to about 14% superior to the existing one known to be effective. [ABSTRACT FROM AUTHOR]

The purpose of this paper is to present some results of the experiments of type simulation embedded in optimization (on model type Systems Dynamics). They allow, not only, the so called, direct optimization, but extended sensitivity analysis of parameters and structures too. Authors have used languages Cosmic and Cosmos to support the learning process in (about) modelled dynamical system. [ABSTRACT FROM AUTHOR]

Efficient locating mechanism is the key issue of establishing peer-to-peer (P2P) resource sharing for mobile Ad Hoc networks (MANET). In this paper, a simple concept of Dynamic Dominant Indexing Set (DDIS) is proposed. By periodically advertising information of shared resources on the backbone of current network topology in a distributed fashion, DDIS gains great improvement on user response time, fault tolerance and scalability. Simulations indicate that DDIS simplifies the searching process and greatly reduces the response delay and thus, it is more appropriate for mobile peer-to-peer source sharing especially in emergent situations. [ABSTRACT FROM AUTHOR]

In this paper, we propose a second-order parameter-uniform convergent hybrid scheme for self-adjoint singular perturbation problems (SPPs) subject to mixed (Robin) type conditions. The cubic spline baesd difference scheme is combined with the classical central difference scheme to obtain monotone scheme. Numerical example is provided to support the theory. Keywords: Finite difference scheme, cubic splines, singular perturbation problems, piece-wise uniform meshes. Subject Classification: AMS 65L10 CR G1.7. [ABSTRACT FROM AUTHOR]

In this paper, new architectures and comprehensive design methodologies of Genetic Algorithms (GAs) based Evolutionally optimized Rule-based Fuzzy Neural Networks (EoRFNN) are introduced and the dynamic search-based GAs is introduced to lead to rapidly optimal convergence over a limited region or a boundary condition. The proposed EoRFNN is based on the Rule-based Fuzzy Neural Networks (RFNN) with the extended structure of fuzzy rules being formed within the networks. In the consequence part of the fuzzy rules, three different forms of the regression polynomials such as constant, linear and modified quadratic are taken into consideration. The structure and parameters of the EoRFNN are optimized by the dynamic search-based GAs. [ABSTRACT FROM AUTHOR]

In this paper, we present a practice-based courseware development methodology for establishing a senior undergraduate network course for the computer-engineering department by reflecting on the rapidly changing information and communication technologies, enforcing practical education, and focusing on the existing and currently used curriculum models. Therefore, we have developed a special method, named it STM (Segmenting, Targeting, and Mapping), and applied it to the courseware development of a practice-based network course. [ABSTRACT FROM AUTHOR]

This paper focuses on the development of an interactive 3D visualization environment for displaying predicted movements by genetic evolution. The integration of interactive 3D visualization and the genetic search approach can relieve the burden of image exploitation tasks occurring due to an explosion of imagery data obtained in battle fields. The visualization component was implemented using In3D - a cross-platform Java class library designed to enable the creation of interactive 3D visualizations of complex information spaces. [ABSTRACT FROM AUTHOR]

This paper presents a first approach to model the communication delays with the aim of improving the design of networked supervisory and control systems (NSCS). Network delays affect the performance of NSCS. Therefore, it is necessary a complete analysis, modeling and simulation of the communication delays upon the connection technology in use (e.g., Ethernet networks). A statistical model is proposed on the basis of autocorrelation of LRD that is imposed by the bursty behavior of the network traffic. [ABSTRACT FROM AUTHOR]

The protein-protein interaction data available today in databases were either determined by experimental methods or predicted by computational methods. Biologists studying a certain mechanism or function are usually interested in a small-scale network of protein-protein interactions related to the mechanism or function of their interest rather than a genome-wide network of protein-protein interactions. We have previously developed a database server that can predict interactions between the proteins submitted by a user. However, the server cannot find proteins from the databases that potentially interact with a protein given by a user. Finding potentially interacting proteins with a protein of interest is more useful than finding interactions between given proteins, but more difficult partly because it involves intensive computation of searching data in the databases and because different databases have different accession numbers and names for a same protein. This paper describes the new online system for predicting interacting proteins with a user-specified human protein and its preliminary results. We believe that this is the first online system for predicting interacting proteins with a given protein and that it is a useful resource for studying protein-protein interactions. [ABSTRACT FROM AUTHOR]

There has been a recent interest in the computational methods for predicting genome-wide protein interactions due to the availability of genome sequences of several species, the difficulty of detecting whole protein interactions in higher species even with the current high-throughput experimental methods, and the common perception of the data generated by high-throughput experimental methods as noisy data. However, data predicted by computational methods as well as that detected by high-throughput experimental methods inherently contain extremely many false positives. Several methods have been developed for estimating the reliability of experimental protein interaction data, but there are few for predicted interaction data. This paper presents a prediction method of protein-protein interactions using the protein domain and sub-cellular localization information, and experimental results of the method to the protein-protein interactions in human. [ABSTRACT FROM AUTHOR]

This paper describes the implementation of a distributed data mining system. The system consists of a web server, a pre-processor for data preparation, a mediator, and agents. A distributed learning algorithm of a decision tree in an agent-mediator communication mechanism is the most important and difficult to achieve the distributed data mining in this system, in the view of implementation. The algorithm has successfully been implemented with several techniques. Its implementation is presented in a UML (Unified Modeling Language) sequence diagram. [ABSTRACT FROM AUTHOR]

This paper proposes a real-time transaction routing algorithm, which allocates real-time transactions to a node in a shared disks cluster. Unlike traditional routing algorithms, which consider load balancing and transaction affinity only, our algorithm also considers transaction priorities inherent to real-time applications. [ABSTRACT FROM AUTHOR]

The applicability and strength of heterogeneous computing systems are derived from their ability to match computing nodes to appropriate tasks since a suite of different machines are interconnected. A good mapping algorithm offers minimal expected completion time and machine idle time. In this paper we propose a new task scheduling algorithm allowing higher performance than the existing algorithms such as the Min-min, Max-min, and Sufferage algorithm. It is achieved by task swapping approach based on the expected completion time and ready time of each machine. Extensive computer simulation validates the proposed algorithm along with the earlier ones. Keywords: Heterogeneous computing, load balancing, task mapping, throughput, scheduling. [ABSTRACT FROM AUTHOR]

For parallel computations requiring massive data Input/Output, one of the goals is to maximize the use of the underling storage topology, in particular exploit the benefit of using local disks. This paper presents a mechanism to distribute independent data records from multiple files to multiple processing nodes and vice-versa. An allocation problem is solved using the Max-Flow algorithm. We give timing results using various read/distribute protocols. One application is to redistributing tasks (regions) for restarting parallel numerical integration runs. [ABSTRACT FROM AUTHOR]

This paper presents the Optimality Theorem in distributed parallel environment. Based on this theorem, a parallel algorithm using network-tree model is presented to compute shortest paths in large-scale networks. The correctness of this algorithm is proved theoretically and a series of computational test problems are performed on PC cluster. Factors such as network size and level of the network, which take effect on the performance, are discussed in detail. Results of the experiments show that the proposed parallel algorithm is efficient in computing shortest paths in large-scale networks, especially when the network size is great. [ABSTRACT FROM AUTHOR]

As the number of Internet services grows, it is becoming more important for the network users to be able to locate and utilize those services that are of interest to them. As a result, a service discovery protocols are coming to play an increasingly important role in highly dynamic networks. In this paper, we describe a new architecture for service discovery called LSD (Lightweight Service Discovery). The LSD is a peer-to-peer cache-based service discovery protocol for ad hoc environments. Through the LSD protocol, small handheld mobile devices with wireless connectivity can detect each other in their neighboring devices, and perceive whether particular shared services are available or not. In addition, the LSD solves the overhead problems which arise when traditional discovery protocols are applied to mobile ad hoc networks. [ABSTRACT FROM AUTHOR]

In this paper, we propose an enhanced location management scheme for minimizing signaling costs in hierarchical Mobile IPv6 (HMIPv6) networks. If the mobile node's mobility is not local, in our proposal, the mobile node sends location update messages to correspondent nodes in the same way as Mobile IPv6 (MIPv6). After the creation of a spatial locality of the mobile node's movement, the mobile node sends location update messages to the correspondent nodes in same way as HMIPv6. Therefore, our proposal can reduce signaling costs in terms of packet transmission delays in HMIPv6 networks. Therefore, our proposal offers considerable performance advantages to MIPv6 and HMIPv6. [ABSTRACT FROM AUTHOR]

In this paper, a new method based on genetic algorithms is proposed for local tuning of fuzzy membership functions. For this purpose, the local adjustment is employed on the initial membership functions. Genetic algorithm is used to investigate discrete points that will be modified on the membership functions Hence, global adjustment does not require and the processing time required for tuning of membership functions is decreased. [ABSTRACT FROM AUTHOR]

The inductive learning algorithms are the very attractive methods generating hypothesis of the target concept on the base of the set of labeled examples. This paper presents some of the rules generation methods, their usefulness for the rule-base classifier and their quality of classification for the medical decision problem. [ABSTRACT FROM AUTHOR]

This paper proposes a new algorithm and an architecture for it to compute the modular multiplication over GF(2m). They are based on the standard basis representation and use the property of irreducible all one polynomial as a modulus. The architecture, named SSM(Semi-Systolic Multiplier) has the critical path with 1-DAND+1-DXOR per cell and the latency of m+1. It has a lower latency and a smaller hardware complexity than previous architectures. Since the proposed architecture has regularity, modularity and concurrency, they are suitable for VLSI implementation. [ABSTRACT FROM AUTHOR]

In this paper we develop a general Homotopy method called the Group Homotopy method to solve the symmetric eigenproblem. The Group Homotopy method overcomes notable drawbacks of the existing Homotopy method, namely, (i) the possibility of breakdown or having a slow rate of convergence in the presence of clustering of the eigenvalues and (ii) the absence of any definite criterion to choose a step size that guarantees the convergence of the method. On the other hand, the Group Homotopy method maintains attractive features of the ordinary Homotopy method such as the natural parallelism and the structure preserving properties. [ABSTRACT FROM AUTHOR]

In this paper the solution of one-phase inverse Stefan problem with Neumann's boundary condition is presented. This problem consists of the reconstruction of the function which describes the heat flux on the boundary, when the position of the moving interface is well-known. The proposed solution is based on the Adomian decomposition method and the least square method. [ABSTRACT FROM AUTHOR]

Georeference is a basic function of remote sensing data processing. Geo-corrected remote sensing data is an important source data for Geographic Information Systems (GIS) and other location services. Large quantity remote sensing data were produced daily by satellites and other sensors. Georeferenceing of these data is time consumable and computationally intensive. To improve efficiency of processing, Grid technologies are applied. This paper focuses on the parallelization of the remote sensing data on a grid platform. According to the features of the algorithm, backwards-decomposition technique is applied to partition MODIS level 1B data. Firstly, partition the output array into evenly sized blocks using regular domain decomposition. Secondly, compute the geographical range of every block. Thirdly, find the GCPs triangulations contained in or intersect with the geographic range. Then extract block from original data in accordance with these triangulations. The extracted block is the data distributed to producer on Grid pool. [ABSTRACT FROM AUTHOR]

In this paper we propose GRAMS, a resource monitoring and analysis system in Grid environment. GRAMS provides an infrastructure for conducting online monitoring and performance analysis of a variety of Grid resources including computational and network devices. Based on analysis on real-time event data as well as historical performance data, steering strategies are given for users or resource scheduler to control the resources. Besides, GRAMS also provides a set of management tools and services portals for user not only to access performance data but also to handle these resources. [ABSTRACT FROM AUTHOR]

The promise of Hive Computing - Organizations being able to acquire all the power they need for only as long as is necessary - is incredibly compelling. Hive Computing has experienced significant success in bringing productivity gains, high performance and cost savings to business applications such as; Enterprise Resource Planning (ERP), Customer Relationship Management (CRM), Supply Chain Management (SCM) and B2B portals (E-commerce sites). Grid Computing has brought productivity gains, high performance and cost savings, but in some places it is largely incomplete, i.e. when it comes to the questions of mission-critical computing in general and transaction oriented in particular. This paper discusses a new approach to the development, deployment and management of mission-critical applications - called Hive Computing - that is designed to complement and extend the vision of Grid Computing. [ABSTRACT FROM AUTHOR]

HLA(High Level Architecture) has been developed to promote interoperability and reusability within the modeling and simulation community. RTI(Run Time Infrastructure) is a software implementation of HLA which is composed of three components: libRTI, FedExec, and RTIExec. However, the previous RTI has an unfriendly execution environment with static and manual resource allocation and execution. In the paper, we present a RTI execution environment using OGSA(Open Grid Service Architecture) which addresses the problems, so called RTI-G. It supports easy-to-useness, transparency and performance by providing users with dynamic resource allocation and automatic execution. Moreover, it provides an unified view of RTI and Grid services by integrating them as web services on OGSA. [ABSTRACT FROM AUTHOR]