Gateway Track

Sessions highlight the success of researchers using a gateway or technology approaches to gateway design, development, and enhancement. Presentations feature the scientific problem, the computational methods and TeraGrid resources used, plus any unique aspects of the gateway that facilitate access to resources. Sessions will feature a variety of technical solutions relevant to gateways, such as fault tolerance, accounting practices, the use of new technologies to reach emerging communities, novel interactive approaches to scheduling, visualization, and more! Many sessions incorporate interactive gateway demonstrations as well as slide presentations to provide a wealth of background information on the project.

Gateway Track Invited Speaker

Bioinformatics gateways to next-generation infrastructures
Christophe Blanchet, Institute of Biology and Chemistry of Proteins, CNRS, Lyon, France
Bioinformatics is now requiring Research Infrastructures able to store very large, complex and heterogeneous biological datasets, and making them available for intensive scientific computing. Facing these growing needs could be done by taking benefits of the new developments in terms of distributed and intensive computing infrastructure like grid or cloud platforms. Defining standards to describe programmatic interfaces for remote biological resources, data and tools, and making them available to scientists through public Web services based on SOAP or REST is also a good way to set up such new distributed Research Infrastructures for Bioinformatics.  From the bioinformatics provider point of view, the challenge is now to succeed the integration of our usual biological data and tools, and to provide users - biologists and bioinformaticians - with efficient interfaces and gateways to these infrastructures. This presentation will detail the developments we are doing about bioinformatics portal and Web services, in the context of the French infrastructure RENABI GRISBI and of the European projects EMBRACE and StratusLab, for example  with the django framework or the S3 cloud storage, to address these needs from our community.

Schedule of Gateway Track Talks

Tuesday August 3rd:

Jim Basney, Von Welch and Nancy Wilkins-Diehr. TeraGrid Science Gateway AAAA Model: Implementation and Lessons Learned

Download the presentation(PPTX)
Abstract: In this paper, we present our experience implementing on the TeraGrid the "Science Gateway AAAA Model" we proposed in our 2005 paper. We describe how we have modified the model based on our experiences, the details of our implementation, an update on the open issues we identified in our paper, and our lessons learned.

John Schmidt and Martin Berzins. Development of the Uintah Gateway for Fluid-Structure-Interaction Problems

Download the presentation(PDF)
Abstract: The Uintah Gateway was designed to allow users to create complex multi-physics Uintah simulations with ease and to run these on large parallel computers. We used the Django web application framework to develop the Uintah Gateway for fluid-structure-interaction problems. We describe using the Gateway from input file creation to data management. We also describe two use cases: one involving a complex fluid-structure interaction problem with multiple simulations and multiple restarts and the other involving novice users just getting started with Uintah. Preliminary results suggest that both novice and advanced productivity increased dramatically using the Gateway.

Greg Johnson, Brandt Westing, Stephen Mock, Maytal Dahan, Matthew Hanlon and Kelly Gaither. The Longhorn Visualization Portal: A Web-Based Interface to the Longhorn Visualization and Data Analysis Cluster
Abstract: On January 4, 2010 the Texas Advanced Computing Center (TACC) at The University of Texas at Austin put into production a Dell/NVIDIA visualization and data analysis cluster, Longhorn, as a TeraGrid resource, the first of the National Science Foundation Office of Cyberinfrastructure eXtreme Digital Visualization machines to be deployed in 2010. Longhorn is configured as a hybrid CPU/GPU system designed to provide remote, interactive visualization and data analysis, and compute-intensive calculations on both CPUs and GPUs. The large per-node memory (48GB on 240 Dell R610 nodes and 144GB on 16 Dell R710 nodes) and powerful GPUs (2 NVIDIA FX 5800s per node) are intended to support serial and parallel visualization and data analysis applications that leverage large memories, multiple graphics processors and multiple computing cores. Longhorn is designed for effective analysis of very large data sets created on TeraGrid HPC systems.

Shaowen Wang and Yan Liu. GISolve 2.0: Geospatial Problem Solving Environment Based on Synthesizing Cyberinfrastructure and Web 2.0
Abstract: Cyberinfrastructure integrates high-performance and distributed computational resources, and represents a promising infrastructure to resolve the challenges involved in computationally intensive and collaborative geospatial problem solving. However, cyberinfrastructure will remain evolving and complex in the foreseeable future for such problem solving to reap its benefits. We present a synthesis approach to simplifying access to cyberinfrastructure by establishing the GISolve Toolkit 2.0.
Specifically, GISolve 2.0 uses Web 2.0 technologies and service-oriented architecture to lower the barriers of accessing cyberinfrastructure for creating user-centric geospatial problem solving environments. Web 2.0 technologies are applied to build highly interactive and collaborative user environments to accommodate highly-usable application and data access within Web browser. By leveraging advanced Web 2.0 user interface and communication technologies (e.g., Yahoo UI, OpenLayers, AJAX), the GISolve 2.0 Web environment (figure 1) provides an integrated interface to generate and access information related to particular problem-solving such as user, application, experiment, computation, data, and visualization. Web browser controls the Web content rendering, user interaction event handling, and data and message communication with backend Web servers. Compared to conventional Web applications which often require frequent Web page refresh and navigation, user experience in GISolve 2.0 is significantly improved to allow most of user interactions upon an analysis to happen within a single Web page without page-refreshing, such as asynchronous Web content fetching, data and parameter validation, drag-and-drop-triggered visualization.
Within the Web-based user environment, the access to cyberinfrastructure resources and services is seamlessly integrated with geospatial problem solving services based on geographic information systems (GIS) and spatial analysis methods and software. Using component-based design, GISolve 2.0 transforms GIS and spatial analysis software into Web services that are interoperable with generic cyberinfrastructure services. To access TeraGrid, the science gateway community account model is implemented as a service that manages gateway credential update and gateway user attribute insertion in job submission. As a new GIS and spatial analysis service is integrated, both Web interface and cyberinfrastructure access component are automatically generated and registered. Consequently, GISolve 2.0 implementation is modular and, thus, allows for flexible deployment and mashup of services by representing every service as a Web resource that can be accessed through standard Web protocols.
Both real and synthetic data within a science application context of discovering geographic patterns of global climate change impact are used to demonstrate the capabilities of GISolve 2.0 that has been deployed as a multi-user geospatial problem solving environment on the National Science Foundation TeraGrid.

Wednesday August 4th:

Wenjun Wu, Thomas Uram, Michael Wilde, Mark Hereld and Michael E. Papka. Accelerating Science Gateway Development with Web 2.0 and Swift
Abstract: A Science Gateway is a computational web portal that includes a community-developed set of tools, applications, and data customized to enable scientists to run scientific simulations, data analysis, and visualization through their web browsers. The major problem of building a science gateway on a Grid environment such as TeraGrid is how to deploy scientific applications rapidly on computational resources and expose these applications as web services to scientists. Although many web-service frameworks have been designed and applied in building domain-specific science gateways, most of these efforts only addressed the issue of adding scientific applications as SOAP services into a service container; they usually don't provide solutions to support web interface generation. Developers still need to spend a lot of time learning web programming to implement a user-friendly and interactive web interface to these services.
In this paper we propose a new application framework that can greatly accelerate the development cycle of science gateway systems. This framework enables science gateway developers to import their domain-specific scientific workflow scripts and generate Web 2.0 gadgets for running these application workflows and visualizing the output from workflow executions without writing any web related code. By assembling these application-specific gadgets and some common gadgets predefined in the framework for workflow management, developers can easily set up a customized computational science gateway to meet community requirements. We demonstrate the utility of the framework with an example from computational biochemistry.

Marlon Pierce, Suresh Marru, Raminder Singh, Archit Kulshrestha and Karthik Muthuraman. Open Grids Computing Environments: Advanced Gateway Support Activities
Abstract: We describe three case studies for providing advanced support for TeraGrid Science Gateways as part of our participation in the Advanced User Support team. These case studies include providing workflow support, robust job management, and mass job submission to existing gateways targeting computational chemistry, biophysics, and bioinformatics, respectively. Selected tools from the Open Grid Computing Environments and other projects were used, demonstrating the need for flexibility when integrating tools from multiple software providers into specific gateways' software stacks.

Thursday August 5th:

Rion Dooley, Maytal Dahan, Stephen Mock and John Boisseau. The TeraGrid Mobile User Portal: Ticketing, Data, and Complete Job Management On The Go
Abstract: The TeraGrid User Portal (TGUP) is a web portal that aggregates and simplifies access to TeraGrid information and services for active TeraGrid users. The purpose of the TGUP is to make using the large number of diverse resources and services of the TeraGrid easier for the national open science community, thus increasing their productivity and the impact of the TeraGrid project. In order to increasing the impact and visibility of the TeraGrid project and the TeraGrid User Portal the team developed an optimized web interface for mobile smartphone users, providing access to time-critical information for TeraGrid users who are away from their workstations.
TGUP Mobile (Mobile) is a lightweight, responsive web application providing a subset of TGUP capabilities via a mobile device. Released in September 2009, Mobile originally offered the ability to view historical job information, read user news items, view TeraGrid resources status, check user allocation information and balances, contact the TeraGrid Help Desk, and view/update user profile information. In its first six months of production operation, TGUP Mobile averaged over thirty-two hundred hits a month and created a new way for users to interact with their science and their TeraGrid. On the heels of this success, Mobile has undergone continual development to provide a broader range of features relevant to mobile users. Recently added features include real time job monitoring across TeraGrid resources, the ability to subscribe for notifications on currently running jobs, and expanded allocation management for Principal Investigators to add and remove users from their allocations. As a result of these additions, usage of the job monitoring portion of Mobile has more than doubled.
In the Spring and Fall of 2010, Mobile will be further updated to provide several long-awaited capabilities. First, in May 2010, Mobile will be updated to add support for expanded user profile information and the ability to browse and respond to consulting tickets. In September of 2010, a fourth update to Mobile will be released adding remote file management and the ability to share data with other TeraGrid users and publicly publish data from TeraGrid resources to the web.
With these new features, the TGUP Mobile application will enable new usage scenarios that meet users at their lifestyle rather than just their laptop. Users will be able to interact with their science seamlessly as they move from their office to the world around them. Future expanded features will allow users to be able to start a job on their workstation before leaving the office. While in commute to a meeting, they will be able to log into Mobile, view the status of the job they just submitted, and subscribe for an email notification when it finishes. After the meeting, they will have received an email on their phone informing them that their job has finished. On their way back to their office, they will be able to log back into Mobile, browse their job data on the remote TeraGrid system and publish their results for the rest of their team too view.
Mobile exists to add value to the TeraGrid user community and expand the capabilities and outreach of the TeraGrid User Portal. User feedback is crucial to the success of Mobile and always appreciated. For more information on Mobile, please visit https://mobile.teragrid.org.

Lan Zhao, Wonjun Lee, Carol Song, Matthew Huber and Aaron Goldner. Bringing High Performance Climate Modeling into the Classroom
Abstract: Climate science educators face great challenges on combining theory with hands-on practices in teaching climate modeling. Typical model runs require large computation and storage resources that may not be available on a campus. Additionally, the training and support required to bring novices up to speed would consume significant class time. The same challenges also exist across many other science and engineering disciplines. The TeraGrid science gateway program is leading the way of a new paradigm in addressing such challenges. As part of the TeraGrid science gateway initiative, The Purdue CCSM portal aims at assisting both research and education users to run Community Climate System Model (CCSM) simulations using the TeraGrid high performance computing resources. It provides a one-stop shop for creating, configuring, running CCSM simulations as well as managing jobs and processing output data. The CCSM portal was used in a Purdue graduate class for students to get hands-on experience with running world class climate simulations and use the results to study climate change impact on political policies. The CCSM portal is based on a service-oriented architecture with multiple interfaces to facilitate training. This paper describes the design of the CCSM portal with the goal of supporting classroom users, the challenges of utilizing the portal in a classroom setting, and the solutions implemented. We present two student projects from the fall 2009 class that successfully used the CCSM portal.

Roberto Barbera, Giuseppe La Rocca, Riccardo Rotondo, Alberto Falzone, Paolo Maggi and Nicola Venuti. Conjugating Science Gateways and Grid Portals into e-Collaboration environments: the LifeRay and GENIUS/EnginFrame use case

Download the presentation(PPT)
Abstract: Nowadays, e.Science uses tools that allow scientists divided by long distances to work on the same project. Among them, Science Gateways are becoming very popular but generally lack the possibility to access directly the grid middleware as Grid Portals do. In the paper we present a new e-Collaboration environment that merges the concepts of Science Gatreway and Grid Portal. An implementation case using LifeRay and GENIUS/Enginframe technologies is shown.

Jeff Wood, Linda Hayden and Raminder Singh. UTILIZING A SOCIAL NETWORKING SITE AS A WEB PORTAL TO PROCESS CRESIS RADAR DATA
Abstract: Scientists have shown that declines in the mass of the Greenland and Antarctic ice sheets will lead to sea level rise affecting large areas of coastlines. Data is being collected by researchers regarding the mass of the ice creating large amounts of data which can be overwhelming to undergraduate and K-12 researchers. Visualization can lead to a better understanding of this data and guide students to involvement in polar research. The goal of this project was the implementation of code to provide an interactive display of ice depth data from Greenland expeditions. The project required the cooperation of the CReSIS and PolarGrid partnerships. The implementation involved the use of a social networking site as a gateway to ice depth data. This project is the basis of a cyberinfrastructure gateway to be implemented at ECSU using an intermediary shared services approach versus the previous generation of component based portals.

Contact Gateways Track Co-Chairs Nancy Wilkins Diehr (SDSC), wilkinsn at sdsc.edu and Maytal Dahan (TACC), maytal at tacc.utexas.edu.