The Gridbus Toolkit: Creating and Managing Utility Grids for eScience and eBusiness
Applications
Level of Tutorial: 25%
Introductory, 25% Intermediate, and 50% Advanced.
Tutorial Speakers:
Dr. Rajkumar Buyya
The University of
Short Abstract:
In spite of a number of
advances in Grid computing, resource management and application scheduling in
such environments continues to be a challenging and complex undertaking. This
is due to geographic distribution of Grid resources owned by different
organizations with different usage policies, cost models and varying load and
availability patterns with time. This tutorial introduces fundamental principles of
Grid computing and computational economy and discusses how they impact on
emerging Computational and Data Grid technologies. It identifies resource
management challenges and presents a Grid Architecture for Computational
Economies that can be realized by leveraging existing technologies. It
then introduces new challenges and requirements introduced by the Grid Economy
on Grid Service Providers (GSPs) and Grid Service
Consumers. We present solutions to these challenges based on our experience in
designing and developing computational and Data Grid technologies such as
Nimrod-G, GridSim, Gridbus
(e.g., Grid Market Directory, Grid Bank, Grid Service Broker, Grid Federation)
and their utilization in driving eScience
and eBusiness applications such as molecular docking,
natural language processing, and portfolio analysis.
Intended Audience:
This tutorial should be of interest to a large number of participants
from academia, government, and commercial organizations as it focuses on both theory and practice of Grid Economy. They
include: (A) students, researchers, and developers interested in creating technologies
and applications for Next Generation Grids with focus on Grid economy, (B)
participants from commercial organizations interested in creating online Grid
marketplace, and (C) users of Grid Computing as we will be offering a live
demonstration of current Grid Economy-based technologies and their
applications during the tutorial.
Background:
We expect participants to have knowledge of Grid computing at the
introductory level. A familiarity of low-level Grid middleware such as Globus
Toolkit will be an advantage.
Extended Abstract:
Grids
aim at exploiting synergies that result from cooperation of autonomous
distributed entities. The synergies that
result for Grid cooperation include the sharing, exchange, selection, and aggregation of geographically
distributed resources (such as computers, data bases, scientific instruments)
for solving large-scale problems in science, engineering, and commerce. For this cooperation to be sustainable, participants
need to have (economic) incentive. Therefore,
“incentive” mechanisms should be considered as one of key design parameters of
Grid architectures.
The Grid community has
embraced the integration of commodity Web services and Grid technologies that
led to the development of Grid services. The recent widespread interested in Grid
computing from commercial organisations is pushing Grid
computing towards mainstream computing and Grid services to become valuable
economic commodities.
In spite of a number of
advances in Grid computing, resource management and scheduling in such environments
continues to be a challenging and complex undertaking. The geographic
distribution of resources owned by different organizations with different usage
policies, cost models and varying load and availability patterns is
problematic. The Grid service
providers (resource owners) and Grid
service consumers (resource users) have different goals, objectives,
strategies, and requirements. To address these resource management challenges,
a distributed computational economy has been recognized as an effective
metaphor for the management of Grid resources as it: (1) enables the regulation
of supply and demand for resources, (2) provides economic incentive for Grid
service providers, and (3) motives the Grid service consumers to trade-off
between deadline, budget, and the required level of quality-of-service. These
factors also promote Grid services to become valuable economic commodities.
This tutorial introduces
fundamental principles of Grid computing and computational economy and
discusses how they impact on emerging Computational and Data Grid technologies.
It identifies resource management challenges and presents a Grid Architecture
for Computational Economies that can be realized by leveraging existing
technologies. It then introduces new challenges and requirements introduced by
the Grid Economy on Grid Service Providers (GSPs) and
Grid Service Consumers. We present solutions to these challenges based on our
experience in designing and developing computational and Data Grid technologies
such as Nimrod-G, GridSim, Gridbus
(e.g., Grid Market Directory, Grid Bank, Grid Service Broker, Grid Federation).
We introduce (a) Grid Market
Directory that allow GSPs to publish their resources
and GSC to discover service providers, (b) different Grid economy models for
resource management, (c) Grid Bank that provides Grid accounting,
authorization, and payment services, (d) Grid Broker that allows users to lease
Grid services at runtime based on their price and users’ QoS
requirements such as the deadline and budget. We present a number of Grid
economy based scheduling algorithms for compute and data intensive applications
on Global Grids.
We demonstrate effectiveness
of Grid economy in resource management by deploying applications such as molecular
docking, natural language processing, and portfolio analysis on our
experimental World-Wide Grid testbed. We also demonstrate
how one can make trade-off QoS requirements such as
the deadline and budget. Results show that the economic algorithms better
utilize computational, storage and network resources than traditional
algorithms.
We conclude the
tutorial by (a) identifying a number of open research topics in Grid computing
with a focus on Grid Economy, (b) discussing our thoughts on new opportunities
for commercial companies to develop a new Grid technologies/products that help
in the realization Grid Exchanges and online Grid marketplaces, and (c) highlighting
sociological and intellectual implications of this new Grid paradigm and its
impact on the computing marketplace.
Outline of the Tutorial:
- Grid Challenges
- Grid Resource Management Challenges
- A Case for Grid Economy
- Service Oriented Grid Architecture and Grid Economy
- Economic Models for Grid Computing
- A Grid Service Publication Directory
- Virtual Organization Services and Grid Economy
- Grid Service Pricing Issues
- Grid Bank and Grid Accounting Services Architecture
- Grid Resource Broker
- Economic Scheduling Algorithms for Computational Grids
- Economic Scheduling Algorithms for Data Grids
- A Case Study in Grid Economy based Systems such as Nimrod-G and Gridbus
- A Case Study in Creation of
Grid Applications from Legacy Software Components
- Demonstration of On Demand Application Deployment and Resource Provisioning within a Grid
Economy Environment
- Advance Reservation and Grid Economy
- Grid Simulator
- Evaluation of Optimisation Strategies
- Performance Metrics
- Simulation and Experimental Results
- Open Research Opportunities
in Grid Economy
- Summary and Concluding
Remarks
About the Speaker:
Rajkumar Buyya is a Senior Lecturer, Storage Technology Corporation Fellow,
and Director of Grid Computing and the Director of Grid Computing and
Distributed Systems Laboratory in the Department of Computer Science and
Software Engineering at the University of Melbourne,
Australia. He received B.E., M.E., and Ph.D. degrees in Computer Science and
Engineering from