| GGF12
Tutorials |
||||||||||||||||||||||||||||||||||||||||||||||||||
|
GGF12 Tutorials Schedule
GGF12 Tutorial Descriptions Instructor: Darren Pulsipher, darrenp@cadence.com Abstract: Cadence Design Systems gridMatrix Grid Initiative Cadence Design Systems has embraced grid technologies, and has dedicated itself to integrating grids into every aspect of their production environment. The Cadence Server Farm Initiative (SFI) has helped reduce development time and costs, both internal and external to Cadence, by integrating grids into their software development and chip design workflow processes. The key grid infrastructure that Cadence developed is called gridMatrix, which allows different organizations to submit workflows seamlessly into the grid. The gridMatrix solution in conjunction with LSF, Sun GridEngine, Condor, common support tools, and standardized deployment practices have turned an integration that initially was not universally embraced into a much anticipated initiative. Cadence has taken a four-pronged approach to deploying a successful grid infrastructure throughout their organization: Standardized Hardware, Common Tools, gridMatrix Infrastructure, and Training. This tutorial will discuss the lessons learned from deploying to over 50 organizations, 17 locations, and over 3000 users around the world. Target Attendess: IT Managers, IT Architects, CIO. [Grid Case
Study 2] Instructors: Roy Reter (NIEHS), Cheryl Doninger (SAS Insitute) Abstract: Life-saving research is a core
competency at the National Institute of Environmental Health Sciences
(NIEHS) based in the United States. NIEHS encountered a dilemma where
data volumes continued to grow at an exponential rate while IT resources
remained limited. Scientists and researchers developed a unique solution
that embraces data mining and grid computing to enhance groundbreaking
research and data analysis into the environmental causes of cancer. This
tutorial shares NIEHS' experience in using a compute grid to distribute
the processing of data and dramatically reduce the elapsed time necessary
to analyze the results. Instructors: John Mincarelli and Sriram Sitaraman (Synopsys Incorporated) Abstract: The deployment of a grid enabled
environment requires detailed attention and several layers of infrastructure
to make an environment easily accessible, usable, productive, intuitive,
and supportable. This tutorial is a case study of the grid enabled environment,
DesignSphere, which was developedand deployed at Synopsys for 5000 engineers
and 15,000 CPU's worldwide. Synopsys has developed a RIDGE framework that
enables Remote access, business Intelligence, Data center planning, Grid
enabled queuing systems, and Environment shaping. The modular components
of RIDGE will be discussed in detail as a case study of what is deployed
at the enterprise level production system at Synopsys and the operational
efficiency that has been achieved by Synopsys by their use. A detailed
financial model that illustrates the operational efficiencies and the
improved efficiency of the computational resources, R&D teams, and
those of IT will also be discussed. [Grid Case
Study 4] Instructor: Tho Ngyuen (SAS Institute) Abstract: The promise of extraordinary increases in computing ability for minimal investment in IT has made grid computing an attractive topic in business, academic and research environments worldwide. With limited IT budgets and increased workload, Texas Tech University (TTU) deployed a compute grid to optimize resources campus-wide. Grid computing leverages unused and underutilized computing resources to attain information that would otherwise take weeks or months to process. TTU has implemented a grid over 200 high powered machines. This tutorial highlights the benefits, implementation approach and lessons learned. [Tutorial
5]: Instructors: Marty Humphrey and Glenn Wasson, University of Virginia Abstract: Web services have become an important component of distributed computing applications over the internet. The merging of grid and web services has lead to the Web Services Resource Framework (WSRF), a series of specifications for performing grid computing on top of web services. WSRF.NET is a project at the University of Virginia that allows the creation of WSRF-compliant web services using the Microsoft .NET platform. This tutorial will provide an overview of important concepts in WSRF, as well as .NET. We will discuss the relationship between the WSRF specifications and current Microsoft technology and the means by which WSRF is rendered on .NET. The WSRF.NET architecture will be described and its programming model explained. We will then cover the process of building services in WSRF.NET via a hands-on lab for directly experimenting with the software. Participants will learn about the WSRF.NET system and how to use it to write WSRF-compliant services and clients. Audience: Attendees interested in WSRF on .NET, building grid services on .NET, and/or inter-operability between .NET and non-.NET systems. Familiarity with both WSRF and .NET is helpful (as we will not be covering .NET concepts beyond those needed to write services in WSRF.NET), but are not required. [Tutorial
6]: Instructor: Lee Liming, Argonne National Laboratory Abstract: While the Globus Toolkit from the Globus Alliance can make Grid projects and products significantly easier, the challenges themselves are far from easy and the Globus Toolkit does not provide a turnkey solution. Success in a Grid project depends on a clear vision of the problems that need to be solved, awareness of existing technologies that can contribute to the solutions (both within and beyond the Globus Toolkit), and a strategy for using the technology to overcome the challenge. This tutorial provides answers to critical questions for Grid project planners and product developers and how the Globus Toolkit fits into a Grid project or product strategy. The Globus Toolkit will be put into context, and examples and roadmaps for the most common uses of the Globus Toolkit will be provided. Audience: IT executives and managers, Project leaders, researchers, scientists, and developers. Attendees will gain a deeper understanding of how the Grid and the Globus Toolkit fit within their plans for producinguseful products and/or applications and for planning successful Grid projects. Attendees should be familiar with the basic principals of information technology. For example: general computer and network architecture, client/server systems, databases, current types of commercial IT products and basic internet concepts. [Tutorial
7] Abstract: Grid portals are gaining popularity among discipline scientists and engineers by simplifying and consolidating access to advanced computing systems and Grid resources. Specialized portals in several fields such as earthquake simulation, meteorological research and space science operations are being used or being developed. Grid portals are also utilized by several Grid projects for management and monitoring of resources and as gateways for Grid users to request and access Grid resources. Recently adopted portlet JSR standard (JCS JSR 168) will increase interoperability of application components (portlets) that mediate interaction between user applications and the web portal framework. In this tutorial two leading portal development projects will describe their approaches to building Grid portals. The emphasis will be on describing the fundamentals of the protlet-based portal architecture and the process of how to build a problem solving application portal. The attendees will learn how to obtain base portal software and tips on customizing them for their own use. The intended audience: Grid practitioners, scientist and engineers using or interested in using Grids, developers of Grid services for user communities. General knowledge of Grid functionality, familiarity with using web-based portals. Attendees will learn the usefulness of Grid Portals in providing easy to use interfaces to Grid resources; how to implement and customize a portal framework as a problem solving gateway for use in a scientific discipline. [Tutorial
8] Instructor: Chris Smith, Platform Computing Abstract: The Community Scheduler Framework
(CSF) is a set of Grid Services, implemented using the Globus Toolkit
3.x, which provides an environment for the development of meta-schedulers
that can dispatch jobs to resource managers such as LSF, SGE or PBS. This
tutorial will provide an orientation of the CSF source code posted on
SourceForge, and will describe the Audience: Developers, system administrators, and grid architects Attendees should have basic familiarity
with the CSF source code posted on SourceForge, an understanding of meta-scheduling
requirements in multi-cluster environments. The tutorial will provide
an orientation of the CSF and will describe the interfaces and mechanism
for extending the CSF functionality. Instructors: Carole Goble (U of Manchester), David De Roure (U of Southhampton), Marlon Pierce (Indiana University) The Semantic Grid Tutorial description: 1. a definition of the Semantic Grid, drawing
a distinction with the Knowledge Grid [Tutorial 10] Instructors: Abstract: This tutorial begins with an overview of
distributed and parallel processing approaches, and introduces DRMAA as
a common API to address the distributed application execution needs of
parallel application developers, system administrators, and end users.
The DRMAA programming interface and job attributes will be detailed, as
well as modes of parallel execution. DRMAA API groups will be discussed
in detail, Existing SGE and Condor implementations as well as the use of the Perl DRMAAc bindings will be demonstrated. A proof-of-concept implementation for the distributed rendering of POVRAY jobs will be developed during the tutorial. Future directions of DRMAA will also be
discussed, including Web Services interface, collaboration with JSDL-WG,
requirements for GRAAP job agreements, etc. |