- 1 CoG Kits Speed Up Development of Data Grid for LIGO
- 2 CoG Kit Helps in Establishing Better Than Real-time Forecast for Mesoscale Weather Events
- 3 Java CoG Kits Transition Grids From Useful to Usable
- 4 Java CoG Kit Used in the Southern California Earthquake Center
- 5 The CoG Kit Team Participated in the Design and Implementation of an Active Chemical Table Server
- 6 A New Era of Validated Chemical Reference Data Has Arrived
- 7 CoG Kits Provide Grid Middleware Compatibility Layer for Portal Designers
- 8 CoG Kit Plays Role in Undergraduate and Graduate Student Education
- 9 Global Grid Forum Analyzes Lessons Learned by CoG Kits
- 10 CoG Kit is Used in Pegasus
CoG Kits Speed Up Development of Data Grid for LIGO
The NSF funded Laser Inferomneter Gravitational-Wave Observatory (LIGO) has used the Python CoG Kit to replicate over 50Tb of data to more then 40 sites on 3 continents. The Python CoG Kit (pyGlobus), developed by DOE’s Lawrence Berkeley National Lab, provides an easy to use interface to the Globus Toolkit from DOE's Argonne National Laboratory. It is currently used by a large number of groups to rapidly develop Grid applications, problem solving environments, and portals. By supporting the rapid development of Grid applications, the Python CoG Kit has allowed LIGO researchers to focus more time on the search for gravitational waves instead of software development. CoG Kit to Help to Improve Access Grid Toolkit The newest generation of the Access Grid software was developed using the Python CoG Kit (pyGlobus). The Access Grid software is an ensemble of resources that enables group-to-group collaborations amongst interdisciplinary scientists. The CoG Kit provides an efficient means for authentication and secure communication between different rooms.
CoG Kit Helps in Establishing Better Than Real-time Forecast for Mesoscale Weather Events
Linked Environments for Atmospheric Discovery.(LEAD) is a large NSF ITR collaboration funded to build the foundation for a cyberinfrastructure capable of doing better than real-time forecasts of mesoscale weather events, such as tornadoes. The research challenge is to create an integrated, scalable framework for identifying, accessing, preparing, assimilating, predicting, managing, analyzing, mining, and visualizing the broad array of meteorological data and model output, independent of format and physical location. The LEAD science gateway is based on the DOE/NSF OGCE portal consortium software and the Java CoG Kit for access to the Grid middleware. It will both advance research on the science of mesoscale weather modeling and provide resources for education. In addition, it is planned to be part of a national network of early-warning crisis management cyberinfrastructure tools. LEAD involves the University of Oklahoma (project leader), NCSA, Indiana University, UCAR, Howard University, the University of Alabama, UNC, and Millersville University.
Java CoG Kits Transition Grids From Useful to Usable
Simon Cox, from the GEODISE project, stated at the last Global Gridforum that the Java CoG Kit is an essential tool to transition the Grid from a "useful" resource to a "usable" resource. GEODISE is a research project hosted in the UK involving multi-disciplinary teams working to build a state-of-the-art engineering design, search, and optimization tool for fluid dynamics. It brings together the collective skills of engineers and computer scientists. It includes intelligent design and search tools that will steer the user through the process of setting up, executing and post-processing design search and optimization activities. Large-scale distributed simulations will be coupled with tools to describe and modify designs, using information from databases and knowledge bases. CoG Kits are helping in this strategy to utilize Grid resources.
Java CoG Kit Used in the Southern California Earthquake Center
Researchers at the annual Southern California Earthquake Center (SCEC) meeting used the Virtual Data Toolkit developed by the the GriPhyN and iVDGL projects to support a seismic analysis. Three different Seismic Hazard-related Calculations (Pathway 1, Pathway 2, and Pathway 4) were created and executed using the SCEC portal, in a secure manner. The system produced workflow templates that identified the necessary computations and their sequence, populated them with data producing the abstract workflows, and then handed those to the GriPhyN Virtual Data System (VDS). Pegasus, one of the VDS components, mapped the workflows consisting of both MPI-based and single processor jobs onto the SCEC and the USC grids. One impressive aspect of the system was that there was a consistent workflow system under different seismic hazard programs. Previously, each pathway ran in an independent and dissimilar manner. The following software was used and contributed to this success: NMI, VDT (including VDS), MCS ( Metadata Catalog Service), Condor, and the Java CoG Kit developed at Argonne National Laboratory. This work represents the collaboration of scientists and developers from SCEC at USC and USC/ISI.
The CoG Kit Team Participated in the Design and Implementation of an Active Chemical Table Server
Chemistry researchers from Argonne National Laboratory and the Multi-Scale Chemical Sciences (CMCS) pilot consulted with the Java CoG Kit team on the design and implementation of a web service that exposes the Active Thermochemical Tables (ATcT) to the community. This service was integrated by the SciDAC CMCS pilot in a convenient Web portal. The ATcT program was used to address the 50 year old question regarding the vaporization enthalpy of graphite, a thermochemical reference value for countless computational and experimental studies as reported in a related success story entitled: "A New Era of Validated Chemical Reference Data Has Arrived." The benefit to the scientists is that this powerful program is now accessible to a number of researchers to conduct similar studies.
A New Era of Validated Chemical Reference Data Has Arrived
The Collaboratory for Multi-Scale Chemical Sciences (CMCS) and the application called Active Thermochemical Tables have ushered in a new era of validated chemical reference data where all the pertinent experiments and computations can be considered by all the experts. A 50 year old question of the vaporization enthalpy of graphite, a thermochemical reference value for countless computational and experimental studies, has been resolved by an IUPAC Task Group empowered by CMCS capabilities. The new value for the enthalpy of formation of carbon atom in the gas phase is now more than twice as accurate. The CMCS Active Thermochemical Tables enabled IUPAC group members to collaborate on the systematic reevaluation of previous experimental data, new state-of-the-art computational chemistry results, and comparison with other data. These new capabilities also enabled a group at DOE’s Argonne National Laboratory to fully confirm a recent revision of the enthalpy of formation of the pivotal combustion and atmospheric radical, hydroxyl (OH), and to further reduce its uncertainty by a factor of ~6.5, thus removing a potential source of uncertainty in current chemical models.
CoG Kits Provide Grid Middleware Compatibility Layer for Portal Designers
The Open Grid Computing Environments (OGCE) actively collaborates with the DOE Portals Consortium, the DOE Portal Services project, and the Java CoG Kit project to develop reusable portal components for building Grid computing portals. OGCE is funded by the NSF's National Middleware Initiative program to provide generic Grid portal systems that may be adapted to specific application portals. The OGCE software release includes contributions of tools developed in part by the DOE Portals Services and the Java CoG kit. DOE Portal Services and OGCE portlets develop sharable and interchangeable components, allowing services such as the Grid Portal Information Repository (GPIR) and audio/video collaboration tools to be disseminated to the larger portal building communities sponsored by the NSF, DOD, and NASA. The OGCE portals release builds on the Java CoG Kit to provide higher level Grid programming abstractions that simplify Grid development and provide interoperability between the OGCE portal software and different Grid toolkit versions. In addition, it provides the ability to stage workflows that are typical as part of the scientific problem solving process. The Java CoG Kit is developed at Argonne National Laboratory.
CoG Kit Plays Role in Undergraduate and Graduate Student Education
Argonne National Laboratory hosted a research experience for undergraduates sponsored by NSF, together with DePaul University and the University of Chicago, during summer 2004. Eight undergraduates and four graduate students were involved in exploring Bioinformatics and Grid computing research issues. The Java CoG Kit was used as one of the elementary teaching tools to access the Grid. The students had the chance to contribute to this effort by developing their own programs. Due to the complexity of the interdisciplinary research, it is important that education of the next generation scientist is conducted as early as possible in the carrier of the students. The Java CoG Kit plays an important role in this aspect because it lets the students focus on science instead of reengineering interfaces to the Grid.
Additionally, the Java CoG Kit team has been invited to participate as part of a graduate operating systems course at Illinois Institute of Technology. From the thirty students, seven selected course projects related to the Java CoG Kit.
In an unrelated news item, four out of thirteen peer-reviewed posters at Supercomputing 2004, in the category Grid Computing, are using Java CoG Kit technologies. Three of these posters have direct participation by researchers of the Java CoG Kit team.
Global Grid Forum Analyzes Lessons Learned by CoG Kits
The Global Grid Forum has instantiated a group to research issues related to defining a simple API for Grid Application developers. Based on the decade long experience of the CoG Kit team in supporting scientific applications, and due to the proven concept of the Java and Python CoG Kits, the CoG Kit PI’s have been invited to join this effort. They participated in invited talks and keynotes in a related workshop. The strong collaboration with this effort is based on the fact that the Java and Python CoG Kits have been identified by the Grid community as an important step towards making the Grid usable. A simple API can lower the barrier into Grid computing.
CoG Kit is Used in Pegasus
The developers of Pegasus, which constitutes a flexible framework to enable the mapping of complex scientific workflows onto the grid, are relying in many ways on the Java CoG Kit to enable the access to remote resources and providing Grid security.