Application Of Grid Computing Research Articles

Hybrid Big Bang-Big Crunch based resource scheduling to improve QoS in cloud infrastructure

The service-oriented computing paradigm changes the way of computing. Emerging technologies like grid computing, cloud computing, and smart health care application have changed the way we compute and communicate. Cloud computing has made computing huge data on the fly and uses flexible resources according to the requirement for real-time applications. Cloud computing comes with pay per use model to pay for only those resources that you have used. Inside the cloud there lie many issues related to efficient and cost-effective models to improve cloud performance and complete the client task with the least cost and high performance. E-Health care services are one of the most computational intensive services in the cloud, they require real-time computing which can only be achieved if the computational resources can compute it in the least time. Cloud can accomplish this using an efficient scheduling algorithm. This manuscript focuses on the task scheduling technique which enhances the performance in real-time with the least execution time, network cost, and execution cost. The presented model is inspired by Big Bang-Big Crunch algorithm in astronomy. The presented algorithm enhances the quality of service by reducing the scheduling delay, network delay with the least resource cost to complete the task in the least cost to the user with high quality of service.

English

Systems now days are requiring huge database and massive power of computation. This cannot be gained using the available computation technology or computers. Execution time and large data processing are problems which are usually encountered in highly demanded applications such as multimedia processing. Grid computing technology offers a possible solution for computational intensive applications. Canny edge detection algorithm is known as the optimum algorithm for edge detection that performs edge detection in five different and computationally extensive stages, which consume large processing time. In this work, grid computing is used to perform Canny edge detection as an application of grid computing in multimedia processing. Univa grid engine is used the middleware for the proposed grid system. It is demonstrated here that the proposed grid based solution that integrates grid computing technology into Canny edge detection reduces the processing time while preserving the expected performance of Canny edge detection. The time reduction factor is about three times for the adopted grid system and may become better with careful selection of the communication networks technology.

Swarm Fuzzy Systems: Knowledge Acquisition in Fuzzy Systems and Its Applications in Grid Computing

This work proposes the use of bio-inspired knowledge acquisition for Fuzzy Systems founded on Swarm Intelligence-Particle Swarm Optimization (SI-PSO). Swarm-based models consider knowledge entities as particles that move in the space to reach the higher quality. Fuzzy Systems following SI-PSO for knowledge acquisition are categorized in this work as Swarm Fuzzy Systems (SFSs). Specifically, two learning methodologies, KASIA (using rule bases as particles in PSO) and KARP (using rules as particles in PSO) are introduced. SFSs performance is studied in a problem of practical importance nowadays with data sets, the learning of fuzzy meta-schedulers in computational grids. Fuzzy meta-schedulers are Fuzzy Systems doing intelligent allocation of jobs to improve the performance of the grid, such as the reduction of the execution time of workload. The scheduling decisions are taken based on the knowledge of the Fuzzy System and in this way, the relevance of their learning process are critical. In this work, compared results of the performance of the different SFSs and a comparison between SFSs and Genetic Fuzzy Systems are presented. Simulations results show that SFSs can achieve a faster convergence and higher quality with a reduced number of control parameters what makes them a good alternative to Genetic Fuzzy Systems.

A LNS-based data placement strategy for data-intensive e-science applications

With the growth of data, the era of big data comes. Big data requires a huge amount of storage and new data placement methods. For data-intensive e-science applications in grid and utility computing, they need to process large data sets resided in different data centres. The data transfer across data centres leads to bandwidth cost and time delay. It is more important to solve the data placement problem for these applications. The existing data placement approaches can only reduce the data transfer frequency, but not improve overall performance. In this paper, we propose a data placement strategy based on large neighbourhood search (LNS). The strategy has initial strategy generation stage and relaxation and re-insertion stage. Our strategy offers the improvements in overall performance, especially in reducing time delay.

Advanced Computer Technologies and Applications in Grid and Cloud Computing

Advanced Computer Technologies and Applications in Grid and Cloud Computing

Optical Networks for Grid and Cloud Computing Applications

The evolution toward grid and cloud computing as observed for over a decennium illustrates the crucial role played by (optical) networks in supporting today's applications. In this paper, we start from an overview of the challenging applications in both academic (further referred to as scientific), enterprise (business) and nonprofessional user (consumer) domains. They pose novel challenges, calling for efficient interworking of IT resources, for both processing and storage, as well as the network that interconnects them and provides access to their users. We outline those novel applications' requirements, including sheer performance attributes (which will determine the quality as perceived by end users of the cloud applications), as well as the ability to adapt to changing demands (usually referred to as elasticity) and possible failures (i.e., resilience). In outlining the foundational concepts that provide the building blocks for grid/cloud solutions that meet the stringent application requirements we highlight, a prominent role is played by optical networking. The pieces of the solution studied in this respect span the optical transport layer as well as mechanisms located in higher layers (e.g., anycast routing, virtualization) and their interworking (e.g., through appropriate control plane extensions and middleware). Based on this study, we conclude by identifying challenges and research opportunities that can enable future-proof optical cloud systems (e.g., pushing the virtualization paradigms to optical networks).

Grid Computing Application for Brain Magnetic Resonance Image Processing

This work emphasizes the use of grid computing and web technology for automatic post-processing of brain magnetic resonance images (MRI) in the context of neuropsychiatric (Alzheimer's disease) research. Post-acquisition image processing is achieved through the interconnection of several individual processes into pipelines. Each process has input and output data ports, options and execution parameters, and performs single tasks such as: a) extracting individual image attributes (e.g. dimensions, orientation, center of mass), b) performing image transformations (e.g. scaling, rotation, skewing, intensity standardization, linear and non-linear registration), c) performing image statistical analyses, and d) producing the necessary quality control images and/or files for user review. The pipelines are built to perform specific sequences of tasks on the alphanumeric data and MRIs contained in our database.The web application is coded in PHP and allows the creation of scripts to create, store and execute pipelines and their instances either on our local cluster or on high-performance computing platforms. To run an instance on an external cluster, the web application opens a communication tunnel through which it copies the necessary files, submits the execution commands and collects the results.We present result on system tests for the processing of a set of 821 brain MRIs from the Alzheimer's Disease Neuroimaging Initiative study via a nonlinear registration pipeline composed of 10 processes. Our results show successful execution on both local and external clusters, and a 4-fold increase in performance if using the external cluster. However, the latter's performance does not scale linearly as queue waiting times and execution overhead increase with the number of tasks to be executed.

Exact and heuristic solution approaches for the Integrated Job Scheduling and Constrained Network Routing Problem

This paper examines the problem of scheduling a number of jobs on a finite set of machines such that the overall profit of executed jobs is maximized. Each job has a certain demand, which must be sent to the executing machine via constrained paths. A job cannot start before all its demands have arrived at the machine. Furthermore, two resource demand transmissions cannot use the same edge in the same time period. The problem has application in grid computing, where a number of geographically distributed machines work together for solving large problems. The machines are connected through an optical network.The problem is formulated as an IP problem and is shown to be NP-hard. An exact solution approach based on Dantzig–Wolfe decomposition is proposed. Also, several heuristic methods are developed by combining heuristics for the job scheduling problem and for the constrained network routing problem.The methods are computationally evaluated on test instances arising from telecommunications with up to 500 jobs and 500 machines. Results show that solving the integrated job scheduling and constrained network routing problem to optimality is very difficult. The exact solution approach performs better than using a standard IP-solver; however, it is still unable to solve several instances. The proposed heuristics generally have good performance. Especially, the First Come First Serve scheduling heuristic combined with a routing strategy, which proposes several good routes for each demand, has good performance with an average solution value gap of 3%. All heuristics have very small running times.

P2P Technology Application and Research of Grid Computing

Grid computing is a new type of network computing platform emerged after the World Wide Web, which objective is to provide a kind of infrastructure that shares all kinds of information for user． This paper brings forward a Grid-P2P model based on Peer-to-Peer computing, and designs a peer-to-peer genetic scheduling algorithm. The simulation results show that this model increased validity and accuracy of grid computing. It can be applied to the optimization of task scheduling successfully.

Approximation Algorithms for Multiprocessor Scheduling under Uncertainty

Motivated by applications in grid computing and project management, we study multiprocessor scheduling in scenarios where there is uncertainty in the successful execution of jobs when assigned to processors. We consider the problem of multiprocessor scheduling under uncertainty, in which we are given n unit-time jobs and m machines, a directed acyclic graph C giving the dependencies among the jobs, and for every job j and machine i, the probability p ij of the successful completion of job j when scheduled on machine i in any given particular step. The goal of the problem is to find a schedule that minimizes the expected makespan, that is, the expected time at which all of the jobs are completed. The problem of multiprocessor scheduling under uncertainty was introduced by Malewicz and was shown to be NP-hard even when all the jobs are independent. In this paper, we present polynomial-time approximation algorithms for the problem, for special cases of the dag C. We obtain an O(log n)-approximation for the case of independent jobs, an O(log mlog nlog (n+m)/log log (n+m))-approximation when C is a collection of disjoint chains, an O(log mlog 2 n)-approximation when C is a collection of directed out- or in-trees, and an O(log mlog 2 nlog (n+m)/log log (n+m))-approximation when C is a directed forest.

Integrating job scheduling and constrained network routing

Abstract This paper examines the NP -hard problem of scheduling jobs on resources such that the overall profit of executed jobs is maximized. Job demand must be sent through a constrained network to the resource before execution can begin. The problem has application in grid computing, where a number of geographically distributed resources connected through an optical network work together for solving large problems. A number of heuristics are proposed along with an exact solution approach based on Dantzig-Wolfe decomposition. The latter has some performance difficulties while the heuristics solve all instances within minutes and with an average solution value gap as low as 3%.

Routing and Wavelength Assignment of Static Manycast Demands Over All-Optical Wavelength-Routed WDM Networks

In this paper we present the static manycast routing and wavelength assignment (MA-RWA) problem along with heuristics and an integer linear program (ILP) to solve it. Manycast is a point-to-multipoint communication paradigm with applications in e-Science, Grid, and cloud computing. A manycast request specifies a candidate set of destinations, of which a subset must be reached. To solve MA-RWA, a light-tree must be assigned to each manycast request in a static set such that the number of wavelengths required is minimized. We present two heuristics, the shortest path heuristic (SPT) and the lambda path heuristic (LPH), a tabu search meta-heuristic (TS), and an ILP formulation. We show that TS provides results close to the optimal solution (from the ILP) for small networks. We then show that TS provides a 10% improvement over LPH and a 30%–40% improvement over SPT for various realistic networks.

The impact of hardware improvement for molecular modeling in a grid environment

Background: Molecular modeling has become an important tool in the process of computational drug discovery. One of the main challenges in the routine application of molecular modeling technique is the excessive computing capability. Recent advances in hardware capability and improvement in software have led to an increasing interest in meeting the demand of massive computing power for molecular modeling. Grid computing can deliver computational as well as large, sustained data-rich and knowledge-intensive resources across distributed heterogeneous sites, which is attracting increasing attention in solving computing intensive problems. Objective: This paper describes the application of grid computing in computational drug discovery. In particular, the impact of hardware improvement for molecular modeling and challenges presented by the grid computing infrastructure domain are discussed. Conclusion: Hardware improvement in grid computing setting can accelerate computational drug discovery at the molecular modeling stage. However, despite the promising results obtained in different projects, there are still some potential problems in the large-scale application of current grid computing techniques to be addressed.

Experimental Investigation of a Peer-to-Peer-Based Architecture for Emerging Consumer Grid Applications

In recent years, the application of grid computing has become more common in the scientific and business communities and is likely to be open to the consumer market in the future. The achievement of this scheme, referred to as consumer grids, requires not only an efficient network paradigm for job transmission, but also a scalable and robust architecture, since there is a large number of dynamic resources and users (perhaps millions) and a high frequency of job requests in the consumer market. It has already been verified that the optical burst switching (OBS) network is a promising candidate for consumer grids. However, all the current architectures proposed for grids over OBS networks are client–server (C/S) based, carrying the disadvantages of poor scalability, poor fault tolerance, and low efficiency with increasing numbers of grid users and jobs. In order to address this issue, a novel peer-to-peer (P2P)-based architecture for optical grids is investigated, and this architecture is experimentally demonstrated on a labeled OBS (LOBS) network testbed. Results from theoretical analysis and experimental demonstration show that the proposed P2P-based architecture is feasible, scalable, and efficient for grid applications. Moreover, it outperforms C/S-based architectures for emerging large-scale consumer-oriented grid applications.

A system-centric scheduling policy for optimizing objectives of application and resource in grid computing

In grid computing, grid users who submit applications and resources providers who provide resources have different motivations when they join the grid. Application-centric scheduling aims to optimize the performance of individual application. Resource-centric scheduling aims to optimize the resource utilization of resources provider. Due to autonomy both in grid users and resource providers, the objectives of application-centric and resource-centric scheduling often conflict. The paper proposes a system-centric scheduling that provides a solution of joint optimization of the objectives for both the grid resource and grid application. Utility functions are used to express the objectives of grid resource and application. The system-centric scheduling policy can be formulated as joint optimization of utilities of grid applications and grid resources, which combine both application centric and resource-centric scheduling benefits. Simulations are conducted to study the performance of the system-centric scheduling algorithm. The experiment results show that the system-centric scheduling algorithm yields significantly better performance than application-centric scheduling algorithm and resource-centric scheduling algorithm.

Dimensions of coupling in middleware

AbstractIt is well accepted that different types of distributed architectures require different degrees of coupling. For example, in client‐server and three‐tier architectures, application components are generally tightly coupled, both with one another and with the underlying middleware. Meanwhile, in off‐line transaction processing, grid computing and mobile applications, the degree of coupling between application components and with the underlying middleware needs to be minimized. Terms such as ‘synchronous’, ‘asynchronous’, ‘blocking’, ‘non‐blocking’, ‘directed’, and ‘non‐directed’ are often used to refer to the degree of coupling required by an architecture or provided by a middleware. However, these terms are used with various connotations. Although various informal definitions have been provided, there is a lack of an overarching formal framework to unambiguously communicate architectural requirements with respect to (de‐)coupling. This article addresses this gap by: (i) formally defining three dimensions of (de‐)coupling; (ii) relating these dimensions to existing middleware; and (iii) proposing notational elements to represent various coupling integration patterns. This article also discusses a prototype that demonstrates the feasibility of its implementation. Copyright © 2009 John Wiley & Sons, Ltd.

An innovative replica consistency decision model with Naive Bayesian Classifier in data grids

Data grid is one of the most popular applications in grid computing. In data grids, replication is a critical technology aiming at lowering the access delay as well as saving network bandwidth by duplicating original data in a distributed manner. Files were considered read‐only in most of the previous research work about data replication. For read/write files, replica consistency is an important issue. In this paper, we propose a novel replica consistency decision model with high adaptability and flexibility using the Naive Bayesian Classifier in order to improve the system performance in data grids. A system prototype has been implemented. The experimental results also prove the excellence of the proposed model.

High-Performance Distributed Video Content Analysis with Parallel-Horus

As the world uses more digital video that requires greater storage space, grid computing is becoming indispensable for urgent problems in multimedia content analysis. Parallel-Horus, a support tool for applications in multimedia grid computing, lets users implement multimedia applications as sequential programs for efficient execution on clusters and grids, based on wide-area multimedia services.

A high-throughput application for the dynamic analysis of structures on a Grid environment

This paper describes a Grid Computing application for the 3D dynamic analysis of large dimension buildings. A previously developed software composed of parallel implementations of eight direct integration methods has been integrated on this application, in order to perform structural simulations on a Grid deployment. The GMarte software abstraction layer has been employed to couple the parallel simulator with the Grid infrastructure. Performing distributed executions has enabled a considerable reduction in the global execution time of structural dynamic studies composed of different design alternatives.

DRAGON: a framework for service provisioning in heterogeneous grid networks

Dynamic resource allocation in GMPLS optical networks (DRAGON) defines a research and experimental framework for high-performance networks required by grid computing and e-science applications. The DRAGON project is developing technology and deploying network infrastructure which allows dynamic provisioning of network resources in order to establish deterministic paths in direct response to end-user requests. This includes multidomain provisioning of traffic-engineering paths using a distributed control plane across heterogeneous network technologies while including mechanisms for authentication, authorization, accounting (AAA), and scheduling. A reference implementation of this framework has been instantiated in the Washington, DC area and is being utilized to conduct research and development into the deployment of optical networks technologies toward the satisfaction of very-high-performance science application requirements.