Computational Grid Resources Research Articles

Analysis of the effect of an external fire on the safety operation of a power plant

The basic aim of this paper is to examine the relevant features of an outdoor fire event and its influence on the surrounding area. The work is based on the analytical study of fire origin, its development and spread. The computer program fire dynamic simulator (FDS) is used to simulate fire behaviour. This program uses large Eddie simulation (LES) to calculate fire development and the spread of combustion products in the environment. The fire source is located in the vicinity of a hazardous plant; e.g., power, chemical, etc. The article presents the brief background of the FDS computer program and the initial and boundary conditions used in the mathematical model. In this analysis, it is often necessary to carry out many repeat simulations of a hazardous event and the computer run time becomes an important factor. Ideally, results should be obtained with a dense computational grid but hardware resources and excessive computational time can preclude this option. The response is to reduce a grid density but to ensure that results are only acceptable if comparable to those obtained with a dense grid. The work presents some corrections of the physical model used and its validation by experimental data, which influence the quality of results.

Optimizing Grid-Based Workflow Execution

Large-scale applications can be expressed as a set of tasks with data dependencies between them, also known as application workflows. Due to the scale and data processing requirements of these applications, they require Grid computing and storage resources. So far, the focus has been on developing easy to use interfaces for composing these workflows and finding an optimal mapping of tasks in the workflow to the Grid resources in order to minimize the completion time of the application. After this mapping is done, a workflow execution engine is required to run the workflow over the mapped resources. In this paper, we show that the performance of the workflow execution engine in executing the workflow can also be a critical factor in determining the workflow completion time. Using Condor as the workflow execution engine, we examine the various factors that affect the completion time of a fine granularity astronomy workflow. We show that changing the system parameters that influence these factors and restructuring the workflow can drastically reduce the completion time of this class of workflows. We also examine the effect on the optimizations developed for the astronomy application on a coarser granularity biology application. We were able to reduce the completion time of the Montage and the Tomography application workflows by 90% and 50%, respectively.

A distributed decomposition policy for computational grid resource allocation optimization based on utility functions

This paper presents a market-oriented resource allocation strategy for grid resource. The proposed model uses the utility functions for calculating the utility of a resource allocation. This allows the integration of different optimization objectives into allocation process. This paper is targeted to solve the above issues by using utility-based optimization scheme. We decompose the optimization problem into two levels of sub problems so that the computational complexity is reduced. Two market levels converge to its optimal points; a globally optimal point is achieved. Total user benefit of the computational grid is maximized when the equilibrium prices are obtained through the service market level optimization and resource market level optimization. The economic model is the basis of an iterative algorithm that, given a finite set of requests, is used to perform optimal resource allocation. The experiments show that scheduling based on pricing directed resource allocation involves less overhead and leads to more efficient resource allocation than conventional Round-Robin scheduling.

Reservation aware operating system for grid economy

Emerging Grid computing, computational resources and instruments are geographically distributed with diverse ownership, introducing a highly dynamic and unpredictable of resource using in Grid environment. Resource management, and scheduling in the Grid environment is complex to undertake. Multi-agent systems and economic model have been proposed to tackle the resource management problem, but little attentions were paid to the underlying technology supporting the scheduling abstractions. This paper proposes a design of reservation aware operating system that able to support the upper level scheduling abstraction in Grid environment. We have developed and evaluated a CPU resource reservation aware system.

Experiences with predicting resource performance on-line in computational grid settings

In this paper, we describe methods for predicting the performance of Computational Grid resources (machines, networks, storage systems, etc.) using computationally inexpensive statistical techniques. The predictions generated in this manner are intended to support adaptive application scheduling in Grid settings, and on-line fault detection. Wedescribe a mixture-of-experts approach to non-parametric, univariate time-series forecasting, and detail the effectiveness of the approach using example data gathered from "production" (i.e. non-experimental) Computational Grid installations.