Skip to main content

2010 | Buch

Grid Resource Management

On-demand Provisioning, Advance Reservation, and Capacity Planning of Grid Resources

verfasst von: Mumtaz Siddiqui, Thomas Fahringer

Verlag: Springer Berlin Heidelberg

Buchreihe : Lecture Notes in Computer Science

insite
SUCHEN

Über dieses Buch

In a dynamic computing environment, such as the Grid, resource management plays a crucial role for making distributed resources available on-demand to anyone from anywhere at any time without undermining the resource autonomy; this becomes an art when dealing with heterogeneous resources distributed under multiple trust domains spanning across the Internet. Today Grid execution environments provide abstract workflow descriptions that need a dynamic mapping to actual deployments; this further accentuates the importance of resource management in the Grid.

This monograph renders boundaries of the Grid resource management, identifies research challenges and proposes new solutions with innovative techniques for on-demand provisioning, automatic deployments, dynamic synthesis, negotiation-based advance reservation and capacity planning of Grid resources. The Grid capacity planning is performed with multi-constrained optimized resource allocations by modelling resource allocation as an on-line strip packing problem and introducing a new solution that optimizes resource utilization and QoS while generating contention-free solutions.

On-demand resource provisioning becomes possible by simplifying abstract resource descriptions independent from the concrete installations. The book further explains the use of the semantic web technologies in the Grid to specify explicit definitions and unambiguous machine interpretable resource descriptions for intelligent resource matching and synthesis; the synthesis process generates new compound resources with aggregated capabilities and prowess. The newly introduced techniques haven been developed and integrated in ASKALON Grid application development and runtime environment, deployed in the Austrian Grid, and demonstrated through well performed experiments.

Inhaltsverzeichnis

Frontmatter

Overview

Frontmatter
1. Introduction
Abstract
A world-wide communication system was developed in the early 1970s to make electronic messaging possible among scientists for the sharing of their research findings with each other. However, the communication protocols were complex and their applications were non-intuitive. Later in the early 1990s, world wide web was created to share not only information but (raw-)data as well in an intuitive way of publishing and sharing instead of messaging. The invention of world wide web has ubiquitized the Internet and enabled scientists to access (raw-)data published by fellow scientists for their own analysis. As a next step, an environment was required to process the huge amounts of data using computers in an efficient way; an environment, in which users and scientists could have communicated and shared not only information and data but resources as well, such as computers and scientific instruments.
Mumtaz Siddiqui, Thomas Fahringer
2. Model
Abstract
This chapter provides an overview of the Grid, its components and discourses with special focus on resource management. It defines important aspects and the technological and architectural advances that have led to the evolution of the Grid thereby setting the foundation for this book.
Mumtaz Siddiqui, Thomas Fahringer

Brokerage

Frontmatter
3. Grid Resource Management and Brokerage System
Mumtaz Siddiqui, Thomas Fahringer
4. Grid Activity Registration, Deployment and Provisioning Framework
Abstract
Resource provisioning is a key concern for implementing an effective resource management as part of the Grid runtime environment; it delivers both physical and logical resources on-demand and shields the application developers from low level details. The previous chapter gives a general overview of the resource management (GridARM) with a detailed description of selection and brokerage problem of physical resources. This chapter introduces GLARE, an integral part of GridARM, that covers logical resources, particularly Grid activities [158]. Existing Grid resource managers concentrate mostly on physical resources. However, some advanced Grid programming environments allow application developers to specify Grid application components (activities) at a higher level of abstraction which then requires an effective mapping between high level resource descriptions i.e. activity types and actual installations i.e. activity deployments This chapter describes GLARE that provides dynamic registration, automatic deployment and on-demand provisioning of activities that can be used to build Grid applications. GLARE simplifies description and representation of both activity types (abstract descriptions) and activity deployments (concrete deployments) so that they can easily be located in the Grid and become available on-demand. GLARE has been implemented as a distributed registry and deployment service by following the superpeer model of GridARM [154].
Mumtaz Siddiqui, Thomas Fahringer

Planning

Frontmatter
5. Allocation Management with Advance Reservation and Service-Level Agreement
Mumtaz Siddiqui, Thomas Fahringer
6. Optimizing Multi-Constrained Allocations with Capacity Planning
Abstract
This chapter introduces capacity planning that exploits advance reservation mechanism. Capacity planning plays a critical role in management of an infrastructure for optimized utilization of perishable resources. This applies to the Grid as well. Once the time has passed the computing power is perished. However, in the Grid, capacity planning is largely ignored due to the dynamic Grid behavior, multi-constrained contending applications, lack of support for advance reservation and its associated challenges like under utilization and agreement enforcement concerns. These issues force a resource manager to make resource allocations at runtime with reduced quality of service (QoS). To remedy these, we introduce Grid capacity planning and management with negotiation-based advance reservation and multi-constrained optimization. A 3-layer negotiation protocol is introduced along with algorithms that optimize resource allocation in order to improve the Grid utility. We model resource allocation as an on-line strip packing problem and introduce a new mechanism that optimizes resource utilization and other QoS parameters while generating contention-free solutions. We have implemented the proposed solution and experimented to demonstrate the effectiveness of our approach.
Mumtaz Siddiqui, Thomas Fahringer

Semantics

Frontmatter
7. Semantics in the Grid: Towards Ontology-Based Resource Provisioning
Mumtaz Siddiqui, Thomas Fahringer
8. Semantics-Based Activity Synthesis: Improving On-Demand Provisioning and Planning
Abstract
In the previous chapter, the possible role of semantics in Grid resource matching and brokerage is discussed. This chapter introduces semantics-based synthesis of activities for automatic workflow generation and improving on-demand resource provisioning. On-demand synthesis of Grid activities plays a significant role in automatic workflow composition and in improving service quality of a Grid resource provisioner. However, in the Grid, synthesis of activities has not been considered due to limited expressiveness of the representation of activity capabilities and the lack of adapted resource management means to take advantage of such activity synthesis. This chapter introduces a new mechanism for automatic synthesis of available activities for the Grid by applying ontology rules. Rule-based synthesis combines multiple primitive activities to form new compound activities. The synthesis process generates new compound activities that can be provisioned as new or alternative options for negotiation and advance reservation. This is a major advantage compared to other approaches that only focus on resource matching. The newly generated compound activities provide aggregated capabilities that otherwise may not be possible; this leads towards an automatic generation of complex workflow applications. Furthermore, we introduce semantics in capacity planning for improving optimization in resource allocation. We demonstrate advantages of semantic-based automatic synthesis of Grid activities.
Mumtaz Siddiqui, Thomas Fahringer

Conclusion

Frontmatter
9. Conclusion
Mumtaz Siddiqui, Thomas Fahringer
Backmatter
Metadaten
Titel
Grid Resource Management
verfasst von
Mumtaz Siddiqui
Thomas Fahringer
Copyright-Jahr
2010
Verlag
Springer Berlin Heidelberg
Electronic ISBN
978-3-642-11579-0
Print ISBN
978-3-642-11578-3
DOI
https://doi.org/10.1007/978-3-642-11579-0