Economic Grid Resource Scheduling Based on Utility Optimization

The aim of this thesis is to design and implement a new Grid Resource Management methodology, where non-massive resources owners would be able to share their resources and integrate human collaboration across multiple domains regardless of network technology, operative platform or administrative domain.<br/>This thesis proposes a distributed and heuristic policy-based resource management architecture for large-scale Grids. The resource management architecture proposed herein is composed of four main building blocs: services management, resource discovery and monitoring, resource scheduling and jobs allocation and activation. The Grid Services Management (GSM) and Jobs Allocation and Activation (JAA) are supported by means of a Policy-based Grid Resource Management Architecture (PbGRMA). This architecture is able to identify service needs arising from diverse sources during the deployment and management of Grid Services, such as requirements demanded by customers, applications and network conditions. Afterwards, the PbGRMA merges these requirements into deployment policies for the corresponding Grid Services. The Grid Resource Discovery and Monitoring (GRDM) is supported by the introduction of the SNMP-based Balanced Load Monitoring Agents for Resource Scheduling (SBLOMARS), in which network and computational resources are monitored by distributed agents. This allows for a flexible, heterogeneous and scalable monitoring system. The Grid Resource Scheduling (GRS) is based on the Balanced Load Multi-Constrained Resource Scheduler (BLOMERS). This heuristic scheduler represents an alternate way of solving the inherent NP-hard problem for resource scheduling in large-scale distributed networks by means of the implementation of a Genetic Algorithm.<br/>Finally, based on the outcome of both the GRDM and GRS, the PbGRMA allocates the corresponding Grid Services by means of its interfaces with Globus ToolKit Middleware and Unix-based CLI commands along of any large-scale Grid Infrastructure. <br/>The synergy obtained by these components allows Grid administrators to exploit the available resources with predetermined levels of Quality of Service (QoS), reducing computational costs and makespan in resource scheduling while ensuring that the resource load is balanced throughout the Grid. The makespan of a schedule is the time required for all jobs to be processed when no one job could be interrupted during its execution and each node can perform at most one operation at any time.<br/>This new approach has been successfully tested in a real large-scale scenario such as Grid5000. The results presented along this Thesis show that our general solution is a reliable, flexible and scalable architecture to deploy and manage Grid Services in large-scale Grid Infrastructures. Moreover, the substitution of the heuristic algorithm approach used into the Grid Resource Scheduling (GRS) phase by other non-heuristics selection algorithms could make our solution useful in smaller Grid Infrastructures.

A distributed multiple dimensional QoS constrained resource scheduling optimization policy in computational grid

\n\t\t\t\t\tThis paper addresses the problem of resource scheduling in a grid computing environment. One of the main goals of grid computing is to share system resources among geographically dispersed users, and schedule resource requests in an efficient manner. Grid computing resources are distributed, heterogeneous, dynamic, and autonomous, which makes resource scheduling a complex problem. This paper proposes a new approach to resource scheduling in grid computing environments, the hierarchical stochastic Petri net (HSPN). The HSPN optimizes grid resource sharing, by categorizing resource requests in three layers, where each layer has special functions for receiving subtasks from, and delivering data to, the layer above or below. We compare the HSPN performance with the Min-min and Max-min resource scheduling algorithms. Our results show that the HSPN performs better than Max-min, but slightly underperforms Min-min.\n\t\t\t\t

The article considers the resource allocation and scheduling problem in a grid computing environment. The article proposes system optimisation scheduling (SOS) that provides a potential solution of joint optimisation of objectives for both the resource and application layer, which combine both application-oriented and resource-oriented scheduling benefits. Grid systems will strive to find an optimal relation between user satisfaction and resource utilisation. Utility functions are used to express grid user's Quality of Service requirement, resource provider's benefit function and system's objectives. In order to verify the efficiency of the proposed scheduling algorithm, we compare the performance of application optimisation scheduling, resource optimisation scheduling, SOS with a traditional Round-Robin algorithm. The simulations study the effect of the request rate and task-to-resource ratio on the different scheduling algorithm.

Grid Computing provides sharing of geographically distributed resources among large scale complex applications. Due to dynamic nature of resources in grid, there is a need of highly efficient job scheduling and resource management policies in grid. In this paper, a novel Grid Resource Scheduler (GRS) is proposed to effectively utilize the available resources in Grid. Proposed GRS contributes, an optimal job scheduling algorithm on Job Rank-Backfilling policy and a resource matching algorithm based on ranking of resources with best fit allocation model. Performance of GRS is measured by considering a web based BLAST algorithm -- a bioinformatics application -- . GRS aims in reducing; Make span of the job workflow and execution time of varied size jobs. The experimental results prove that proposed job scheduling and resource matching algorithms performs better when evaluated against widely used FCFS job scheduling and biggest free resource matching strategies.

Performance evaluation of grid resource management and scheduling system is of key importance to the success of the grid. In this paper, we introduce a generalized stochastic Petri net (GSPN) model for such purpose. The model is simple with clarity and can be applied to other research issues on grid to provide them with performance evaluation capabilities.

In order to resolve resource management and scheduling problem in Manufacturing Grid (MG) – an application of Grid technology, we develop a resource management and scheduling system with the interaction of Manufacturing Grid Information Service (MGIS) and the Manufacturing Grid Resource Scheduler (MGRS). The former, MGIS, provides fundamental mechanisms for remote resource encapsulating, registration, and monitoring, and the latter, MGRS, performs scheduling roles as Global Process Planning (GPP) analyzing, resource discovery, resource selection, and resource mapping.

Task completion time and cost are two significant criteria for the selection of manipulator system. For a given task, several Pareto solutions of manipulator systems should be derived based on the evaluation of these two criteria. However, this process requires a large calculation time. In this paper, we propose a method that can select appropriate systems by evaluating task completion time and cost within the desired calculation time. In the proposed method, multiple objective particle swarm optimization (MOPSO) is employed to search for appropriate manipulator systems from a set of candidate systems. Location optimization and motion coordination are integrated to derive the task completion time and the relative cost is used to evaluate the cost of a manipulator system. We employ particle swarm optimization (PSO) for location optimization and use nearest-neighborhood algorithm (NNA) for motion coordination, since PSO and NNA have a high speed of convergence to a good solution. The proposed method is applied to a set of tasks and is proved to be effective and practical.

Improving the grid resource matching performance under large-scale users and computing resources environment is an important and challenging problem. Many existing grid resource matching and scheduling algorithms need search and compare every grid computing resource node without considering characteristics of grid resource and users' behavior. At the same time, with service oriented grid development, grid system should provide personalized service in scheduling process. Therefore, this paper proposes a novel grid resource matching algorithm based improved collaborative filtering algorithm that can give user recommendation resource set by computing the best approximate resource based on user's behavior.

Selecting appropriate resources for running a job efficiently is one of the common objectives in a computational grid. Resource scheduling should consider the specific characteristics of the application, and decide the metrics to be used accordingly. This paper presents a distributed resource scheduling framework mainly consisting of a job scheduler and a local scheduler. In order to meet the requirements of different applications, we adopt HGSA, a Heuristic-based Greedy Scheduling Algorithm, to schedule jobs in the grid, where the heuristic knowledge is the metric weights of the computing resources and the metric workload impact factors. The metric weight is used to control the effect of the metric on the application. For different applications, only metric weights and the metric workload impact factors need to be changed, while the scheduling algorithm remains the same. Experimental results are presented to demonstrate the adaptability of the HGSA.

Grid resource management and scheduling are confronted with a great challenge. This paper presented a dynamic resource price-adjusting (RPA) strategy in computational Grid. Three new concepts and two new evaluation standards are defined. On the basis of these theories, a Grid resource scheduling algorithm based on marker-driven using dynamic RPA strategy is proposed. With the help of GridSim toolkit, the simulation of resource scheduling algorithm proposed in the paper is realized, and the new algorithm is compared with GridSim DBC (deadline and budget constrained) scheduling algorithms. The experimental results show that the new algorithm improves the performance of Grid in aspect of increasing the task accomplishment ratio and solving the problem of load imbalance.

To enable service level subscription in computing grid, the economic concept of differentiated marketing is introduced into grid resource scheduling and a novel resource-scheduling model, Differentiated Service Model (DSM), is presented. DSM uses the Strictness Factor, Greediness Factor and Credit Score to describe the characteristics of the users and resources, like service requirement and service level. A series of related strategies, including resource evaluating, job risk exposure rating, operation adjustment etc., have been developed to conduct the market players' behavior. A simulation examined DSM's performance and showed its effectiveness: With its aggregate job delivery punctuality being approximate to that of the conventional resource scheduling model, DSM delivered services to users with different quality corresponding to the subscribed service level, especially under a heavy system workload; DSM also balanced the benefit distribution among resources, avoiding the situation that some particular resources monopolize the market share.

The accelerated development in peer‐to‐peer and Grid computing has positioned them as promising next‐generation computing platforms. They enable the creation of virtual enterprises for sharing resources distributed across the world. However, resource management, application development and usage models in these environments is a complex undertaking. This is due to the geographic distribution of resources that are owned by different organizations or peers. The resource owners of each of these resources have different usage or access policies and cost models, and varying loads and availability. In order to address complex resource management issues, we have proposed a computational economy framework for resource allocation and for regulating supply and demand in Grid computing environments. This framework provides mechanisms for optimizing resource provider and consumer objective functions through trading and brokering services. In a real world market, there exist various economic models for setting the price of services based on supply‐and‐demand and their value to the user. They include commodity market, posted price, tender and auction models. In this paper, we discuss the use of these models for interaction between Grid components to decide resource service value, and the necessary infrastructure to realize each model. In addition to usual services offered by Grid computing systems, we need an infrastructure to support interaction protocols, allocation mechanisms, currency, secure banking and enforcement services. We briefly discuss existing technologies that provide some of these services and show their usage in developing the Nimrod‐G grid resource broker. Furthermore, we demonstrate the effectiveness of some of the economic models in resource trading and scheduling using the Nimrod/G resource broker, with deadline and cost constrained scheduling for two different optimization strategies, on the World‐Wide Grid testbed that has resources distributed across five continents. Copyright © 2002 John Wiley & Sons, Ltd.

In order to solve the problem of slow resource scheduling in the process of smart city management, the author proposes a smart city information extraction and data planning system based on the Internet of Things. IoT requires different technologies for analysis for different data types. Managers use different IoT applications to analyze data from different devices and integrate relevant data for possible machine failure or emergency in smart home applications. Situation is predicted. The system includes a cloud platform intelligence center, uses the cloud management module to monitor various hardware devices, constructs the cloud computing resource scheduling objective function, uses the cultural particle swarm algorithm to solve the objective function, and obtains the cloud computing resource scheduling scheme. The experimental results show that the overall utilization rate of the system is the best, close to 100%. Conclusion. The system can realize the effective management of the smart city and monitor the city situation in real time. When implementing resource scheduling, the task completion time is short, the system utilization rate is high, and the resources can be maximized.

Deep learning (DL) is gaining popularity in various fields, such as computer vision, speech identification, and self-driving vehicles. Blackwell graphical processing units (GPUs) and Grace central processing units (CPUs) are widely utilized in computing platforms, including portable gadgets and supercomputers. Optimizing performance for multiple-programming workloads on such devices is crucial, as a single application cannot typically utilize all available resources. Developing effective GPU resource administration and scheduling systems is crucial for effective distribution among several clients. Nevertheless, current versions have several flaws. Clients often provide wrong task resource requirements, resulting in cluster resource low utilization. DL-based resource scheduling without considering cluster network features leads to poor operational efficiency. To address these difficulties, this study develops a distributed data center (DDC) framework using Blackwell GPUs and Grace CPUs for resource scheduling and accelerated computing with DL. This study offers a regression model-based technique for estimating task resource requirements, preventing clients from overallocating computer resources. This paper demonstrates smart cluster network-effective scheduling approaches for both prompt and batch forms based on the resource need assessment methodologies described. Study findings demonstrate that the suggested system can improve task computation by 19% overall and reduce mean job completion duration (JCD) by 22% when evaluated against modern approaches. Furthermore, the presented optimization approaches are useful in a variety of contexts.