8:30 am–12:00 noon
1:30 pm–5:00 pm
The TeraGrid '06 Conference includes half-day and full-day tutorials prior to the start of the conference. The morning session will run from 8:30am until 12:00 noon. The afternoon session will run from 1:30pm until 5:00pm. These tutorials are intended to provide participants with opportunities to explore topics in depth. Many of the tutorials will include hands-on activities with the participants. Please register for the tutorials you plan to attend.
The conference will have a limited number of computers available for the hands-on sessions, and we would very much appreciate your being able to bring your own laptop computer to the tutorials so that you can participate in the hands-on activities. Requirements for any software that may need to be pre-loaded will be sent to people registered for the tutorials in advance of the conference.
In this tutorial we will provide an overview of the visualization hardware resources available on the TeraGrid. We will describe the visualization services that are currently available from each of the TeraGrid resource partners. We will provide an introduction to the UC/ANL ParaView portal, the TACC VNC-based ParaView portal, and the Purdue distributed rendering environment. It is our goal to make these demonstrations as hands-on as possible. We will also discuss our vision for a more integrated TeraGrid visualization environment moving forward. The session will conclude with an open discussion where users are encouraged to share their views on further visualization services that are needed.
For the hands-on portion of the tutorial participants will need a laptop with a web browser. The following software is also needed:
TGviz Gateway
TACC VNC-based ParaView portal
Purdue TeraDRE
See the TeraGrid documentation on creating an X.509 certificate if you have a TeraGrid account, but do not have an X.509 certificate or need to be added to a grid-mapfile.
This tutorial is aimed at current and prospective users of the TeraGrid. A team of infrastructure experts and user consultants will present the various resources, the basic methods for accessing them, and a set of application scenarios based on scientific projects that have already used these resources. There will be hands-on work opportunities, so we strongly recommend that participants bring their laptops.
The tutorial content is oriented to participants that may be beginners with respect to TeraGrid, but they should know something about Unix, HPC, and some high-level language. For those who just want to use portals without any HPC knowledge, you may want to consider attending the TG Science Gateways tutorial.
Software requirements for this tutorial include:
8:30 AM
Introduction to TeraGrid Resources and Services
Sergiu Sanielevici, Jim Marsteller
9:15 AM
Initiating TeraGrid Sessions
Eric Roberts, Raghu Reddy
9:45 AM
Towards the Petascale: Massively Parallel Computing
Raghu Reddy
10:15 AM
Break
10:30 AM
TeraGrid Roaming
Jay Alameda
10:50 AM
Parameter Sweeps: GridShell
Ed Walker
11:20 AM
Cross Site Data Transfers
Krishna Muriki
12:00 PM
Lunch
1:30 PM
Implementing MPI for better performance and scalability
Byoung-Do Kim
2:00 PM
Distributed Parallel Processing: mpich-G2
Nick Karonis, Krishna Muriki
2:15 PM
Distributed Parallel Processing: mpich-VMI
Avneesh Pant
2:30 PM
Data Collections & Data management with SRB
Roman Olschanowsky
3:15 PM
Break
3:30 PM
Client job submission/tracking and data transfers in GridChem
Kent Milfeld
4:00 PM
TeraGrid Visualization
Kelly Gaither
4:45 PM
Wrap-Up Q&A with All Presenters
5:00 PM
Adjourn
This full-day tutorial provides participants with four perspectives on integrating cyberinfrastructure into learning opportunities. Following the four sessions, there will be an open discussion of these and other methods and practices that provide good practices for engaging and retaining today's youth in science, technology, engineering and mathematics.
We encourage all participants to bring their own laptops for various hands-on activities that will occur during this tutorial.
Software requirements for this tutorial include:
8:30 AM
Welcome
Edee Wiziecki, NCSA
8:35 AM
e-Labs—a Collaborative Learning Environment
Marge Bardeen, Fermi National Laboratory
Thomas Jordan, Fermi National Laboratory, jordant@fnal.gov
Explore a new collaborative learning environment that uses grid computing techniques to support high school student investigations. In this first e-Lab students use a classroom detector to gather and subsequently upload those data to an e-Lab website. They use a web-browser to explore the data using our custom interface to Griphyn's Virtual Data Toolkit. Students apply techniques such as virtual data transformations, workflows, metadata cataloging and indexing, data product provenance and persistence. They use job planners for execution locally and on the grid. Students also share results in the form of online posters and ask each other questions in this asynchronous environment. Students can discover and extend the research of other students, modeling the processes of modern large-scale scientific collaborations. Also, the e-Lab provides tools for teachers to guide student work throughout an investigation.
10:00 AM
Break
10:30 AM
NCN / nanoHUB.org: A fully operational Cyberinfrastructure that is transforming how students learn, teachers teach, and researchers & engineers work
Gerhard Klimeck1, Mark Lundstrom1, Michael McLennan2, Sebastien Goasguen2, Krishna P.C. Madhavan2
1 School of Electrical and Computer Engineering
2 Information Technology at Purdue (ITaP) Network for Computational Nanotechnology, Purdue University, West Lafayette, IN-47907, USA.
The Network for Computational Nanotechnology (NCN) was established as a multi-university initiative to create a community resource for nanoscience and nanotechnology—online services for research, education and collaboration. The NCN's cyberinfrastructure delivers online simulation, courses, tutorials, services for collaboration, and more. The NCN's objective is to put theory, simulation, experimental and computational research, and education together with cyberinfrastructure in a way that empowers the development of nanotechnology as a new engineering discipline. NCN's central outreach vehicle is the community web site nanoHUB.org.
The nanoHUB provides unique educational resources, collaborative services, and delivers simulation, visualization, and high-performance computing services online free of charge. The nanoHUB's signature service is online simulation. The power of simulation is fully realized, when software leaves the domain of the computational experts and is released to users with real problems to solve. Acquisition, installation, and maintainance of software is, however, a barrier to its use—especially in new fields where commercial packages are not available. The NCN's goal is to dramatically lower the barrier to the pervasive use of simulations in research and education.
The 3-D Nanoelectronic Modeling Tool (NEMO 3-D) is an electronic structure simulation code for the analysis of quantum dots, quantum wells, nanowires, and impurities. NEMO 3-D uses the Valence Force Field (VFF) method for strain and the empirical tight binding (ETB) for the electronic structure calculations. Various ETB models are available, ranging from single s orbitals (single band effective mass), over sp3s* to sp3d5s* models, with and without explicit representation of spin. The code is highly optimized for operation on cluster computing systems. Simulations of systems of 64 million atoms (strain) and 21 million atoms have been demonstrated on TeraGrid systems. Such calculations require around 12 hours of compute time on 64 CPUs. Significantly simplified simulations that are geared for educational purposes that utilize a single s orbital model require only a few seconds on a single CPU. NEMO 3-D therefore offers the opportunity to engage both educators and advanced researchers, utilizing a single code. An educational version of NEMO 3-D has been release on the nanoHUB and user-friendly versions that enable large-scale parallel executions are under development.
The presentation reviews the mission of the NCN exemplified by the development and deployment of the NEMO 3-D tool. A brief introduction to quantum dots and simulations with NEMO 3-D will be given.
This work is supported by grants of the Army Research Office, Laboratory for Physical Science, and the NSF under grant EEC-0228390.
Noon
Lunch
1:30 PM
Research Experiences Brought to the Museum Floor
Daniela Rosner, Adler Planetarium & Astronomy Museum, drosner@adlerplanetarium.org
In this tutorial we will explore the range of projects that the Adler Planetarium & Astronomy Museum has created for visitors to engage in authentic astronomy research experiences. These projects have led to the development of the cosmic ray i-Lab, a collaborative lab that is part of the grid-enabled Interactions in Understanding the Universe education and outreach project. The i-Lab will bring an appreciation of scientific inquiry to a wide informal education audience by allowing museum visitors to investigate cosmic ray physics. The Adler is in the vanguard of science centers that use cyberinfrastructure to initiate public participation and investigation in the physical sciences. The Adler has developed cyberinfrastructure to involve visitors in simple research experiences and produced tools that use the National Virtual Observatory to explore complex science topics. The tutorial will include a demonstration of the i-Lab as well as demonstrations of some of these other cyberinfrastructure initiatives.
2:30 PM
Computational Chemistry Grid (CCG): A Production Cyber Infrastructure for Computational Chemistry
Leslie Southern, Ohio Supercomputer Center, leslie@osc.edu
Jim Giuliani, Ohio Supercomputer Center, jimg@osc.edu
The Computational Chemistry Grid (CCG) is a 3-year, National Middleware Initiative (NMI) program to develop Cyberinfrastructure for the chemistry community. The CCG is led by the University of Kentucky, and involves collaborating sites at Louisiana State University, the Ohio Supercomputer Center, Texas Advanced Computing Center, and the National Center for Supercomputing Applications. The CCG provides access to high performance computing resources for computational chemistry through a desktop client. The desktop client relies only on standard java-based grid technologies, and is an environment tailored for building inputs, submitting jobs, and post processing data for specific applications. The WSRF grid services function as an interface between the client and site resources of the virtual organization and support authentication, reliable file transfer, job submission, accounting and monitoring. Combined, the CCG offers a streamlined interface to remote site resources and provides robust services to users who need one stop access to multiple sites from a client application development environment.
3:30 PM
Break
4:00 PM
Open discussion and exchange of ideas on successful strategies for engaging and retaining today's youth in science, technology, engineering and mathematics
The Virtual Data System (VDS) is a toolkit for expressing, running, and tracking scientific workflows in Grid environments. VDS workflows can range in scale from desktop analyses to massive-scale Grid computations.
With VDS, users can use the VDL workflow specification language to describe abstract, location-independent workflow recipes for computing scientific datasets. From these abstract descriptions, Pegasus, the VDS planner, automates workflow generation and enables the use of TeraGrid, Open Science Grid, and other resources.
Using VDS to run applications on the TeraGrid automatically records the provenance of derived data, and enables users to plan and track the computational workflows required to derive a particular data product, share workflow specifications with colleagues, and locate computational procedures with desired characteristics.
This tutorial describes the foundations of the virtual data concept, and presents a practitioner-oriented introduction to the Grid-based virtual data tools. Case studies of using virtual data in computing problems in high-energy physics, biology, medical research, astronomy and astrophysics will be presented.
Software requirements for this tutorial include:
The TeraGrid Science Gateways program enables entire communities of users to use the national resources through common interfaces and tools supporting data sets and applications that are important to the shared scientific goals of the community. Examples of different types of Gateways include:
This tutorial includes a brief overview of the TeraGrid Science Gateways projects resources and environment and configuration and descriptions of available services. A requirements analysis of current Science Gateways is driving a set of common service interfaces, policies, and constructs that will be used to facilitate science gateways. These common requirements will be discussed as background, with the primary focus of the tutorial providing an overview of specific steps involved in deploying a Science Gateway to provide access to TeraGrid for a scientific community. Topics will include recommendations for Gateway design to streamline interfacing with the TeraGrid and also related policies and services that are necessary for deployment and support of a TeraGrid Science Gateway.
The programming techniques learned in this tutorial will be applicable in many grid communities. Attendees can expect to learn practical principles relating to Grid security, Web services, portal technologies, and techniques for adapting existing community portals to take advantage of TeraGrid services and resources. Several working Science Gateways and associated applications will be used as examples to illustrate these capabilities. These include:
Software requirements for this tutorial include:
LEAD
GISolve
RENCI
nanoHUB
HDF5 Tutorial is a half day tutorial and consists of two parts: Introduction to HDF5 and Advanced Features of HDF5. Details of the HDF5 software can be viewed at http://hdf.ncsa.uiuc.edu/HDF5.
The tutorial is structured in a way that will benefit both beginners and advanced HDF5 users. HDF team will provide guests logins on the HDF systems for hands-on sessions.
The introductory portion will cover basic HDF5 data model and explain the advantages of using the HDF5 software in the TeraGrid environment. We will give an overview of the HDF5 libraries and tools, and discuss the HDF5 programming model. Simple C and Fortran examples will be used to illustrate HDF5 concepts.
The advanced portion will cover advanced features of the HDF5 library for achieving better I/O performance and efficient storage. The following HDF5 features will be discussed: partial I/O, compression and other filters including n-bit and scale+offset filters, and data storage options. Significant time will be devoted to the discussion of complex HDF5 datatypes such as strings, variable-length, array and compound datatypes.
Both sessions will work with the Tutorial examples and exercises during the hands-on portion.
Software requirements for this tutorial include:
What is a TeraGrid data collection? What does it mean to make data sets available on TeraGrid? What data repositories are available? How to access, manage and publish TeraGrid data collections? What is a data grid and how to use it? What are the usage models for TeraGrid data collections whether they are tied to other TeraGrid resources or not? What are the storage and allocations guidelines and policies in conjunction with data collections?
There is an ongoing Data Collections Requirements Analysis Team (RAT) and well as the Data Collections working group whose outcome will be presented during this tutorial.
Software requirements for this tutorial include:
This instructor-lead, half-day tutorial provides an introduction to programming Web services in the Java language for use on TeraGrid. The tutorial teaches developers how to build a Java Service that makes use of GT4 mechanisms for state management, security, registry and related topics. GT4 is open source middleware that is used in building grids around the world, including TeraGrid. The tutorial is organized in two sections: an introductory lecture and a series of self-directed hands-on exercises in which students add increasing functionality to a skeletal service implementation. Fundamental patterns and interactions of grid computing are highlighted.
The tutorial begins with an instructor-led lecture that introduces key concepts, followed by a series of self-directed hands-on exercises in which attendees add increasing functionality to a skeletal service implementation. Registered attendees will receive instructions prior to the conference on what the hardware/software requirements for their own equipment is, and systems that meet the requirements will be provided for attendees who do not have such equipment.
The tutorial assumes each attendee has access to a network-enabled computer that is pre-loaded with a small set of open-source software. Attendees must be able to run all the software listed in the prerequisites in order to participate in the tutorial. Software requirements for this tutorial include:
Note! It is the attendees' responsibility to insure that their networking, ant and jdk are configured and working properly prior to the tutorial. The integrity of ant/jdk installations can be verified by building this sample code:
The TeraGrid project is funded by the National Science Foundation and includes eight partners:
NCSA, SDSC, Argonne, PSC, ORNL, Purdue, Indiana, and TACC.
Please email the help@teragrid.org with questions or comments.
This site is XHTML 1.0 Transitional, CSS, & Section 508 compliant.