Session on Innovative and Collaborative Problem Solving Environment (PSE) in Distributed Resources Contact: S. KAWATA Schedule: W15 Problem Solving Environment (PSE) is a newly emerging scientific and technological active area in eScience. PSEs provide innovative computational facilities for easy incorporation of novel solution methods to solve a target class of problems in Grid environments, distributed and heterogeneous resources, collaborative environments and so on. Key issues addressed in this Session include PSE for Grid, PSE for collaborations, PSE for heterogeneous distributed system management, PSE for application developments, PSE for scientific computing and PSE for education, as well as PSEs for eSceince-relating issues. PSEs provide computational facilities to solve target problems in the novel way in the fields presented above. Grid and distributed systems are complicated, and contain huge information systems including heterogeneous computer hardware resources, application softwares, middlewares, experimental instruments, and so on. The PSEs are facilities to help users work on the complicated systems. The PSE researches and technology have been intensively explored, and at present the PSEs have started to realize this PSE dream: for example, PSE for Grid deploys middleware or application software on heterogeneous distributed resources across multi sites, and PSEs for application developments may generate, for example, parallel software. The PSEs are also essentially important for the Grid projects. Papers relating to these PSEs in eSceince are welcome to the PSE Session. http://www.ee.utsunomiya-u.ac.jp/~kawatalab/pse/workshop/i2006/e-schedule2006.html
Scientific Workflows and Business workflow standards in e-Science Contact: A.S.Z. Belloum Schedule: W21 In e-Science environments, Scientific Workflow Management Systems (SWMS) hide the integration details at different layers of middleware such as for managing Grid resources, computing tasks, data and information, and automate the management of experiment routines. The workshop focuses on practical aspects of utilising workflow techniques to fill the gap between the e-Science applications on one hand and the middleware (Grid) and the low level infrastructure on the other hand. Recently development in Grid technology have shown a convergence between business workflow standards such as BPEL4WS and scientific workflows and scientific workflow management systems: design, implementation, applications in all fields of computational science, interoperability among workflows and the e-Science infrastructure, e.g., knowledge framework, for workflow management.
The workshop invites well known researchers in the domain of workflow systems from both academia and industry to present the last research outcome in using workflow systems and standards to tackle e-Science issues http://www.science.uva.nl/~adam/workshop/vl-e-workshop.htm
Running Production Grids Contact: A. Hasan Schedule: W23 This workshop intends publicise the tools/approaches that people have written to enhance their production grids as well as publicise problems experienced when running production grids. Such tools/approaches range from troubleshooting tools to debug problems when running on the grid, monitoring tools to ensure the systems are functioning properly and to help in debugging problems. Tools/approaches to deploy systems, to upgrade systems, to schedule outages etc are also necessary parts of a production system. It would also be very useful to collect together production grids experience what tools are needed to ensure a more robust system.
Biologically-inspired Optimisation Methods for Parallel and Distributed Architectures: Algorithms, Systems and Applications Contact: A. Lewis Schedule: W24 This workshop invites papers discussing recent advances in the development and application of biologically-inspired optimisation algorithms to the field of computational science. We encourage submission of papers describing new concepts and strategies, and systems and tools providing practical implementations, including hardware and software aspects. Of particular interest are new approaches in multi-objective optimisation and optimisation in dynamic environments. In addition, we are interested in application papers discussing the power and applicability of these novel methods to real-world problems in both well-established areas, such as computational engineering, and emerging fields such as computational biology. http://eresearch.griffith.edu.au/workshops/eScience
Open Grid Forum Contact: C.T.A.M. de Laat Schedule: W27 In this workshop leaders from the OGF (merger of GGF and EGA) community will present roadmaps, strategies and recent advances on important standards developments in the grid community with application in e-Science. http://www.ggf.org
Collaborative Remote Laboratories (CRL 2006) Contact: W.H. Schiffmann Schedule: W28 eScience is characterized by the cooperation of distributed research groups who share data and powerful computing environments. Often immense data sets that were produced by expensive equipment need to be accessed and evaluated by collaborating research groups who are working at distant locations. In order to reduce the amount of experimental data, remote control of the experimental setup should be provided. Thus, remotely controllable laboratories are necessary to conduct measurements on demand and with respect to specific scientific problems. Web-Platforms for mutual exclusive access are required to grant just one experimenter the control of the laboratory's equipment at a particular time while allowing others to follow the experiment. The acquired data must be made accessible to collaborating researchers and simultaneously a communication infrastructure for the joint evaluation of these measurements is needed. The objective of the workshop is to provide a forum for the presentation of new ideas to address the above requirements, which are essential to build collaborative remote laboratories. http://ra.fernuni-hagen.de/downloads/CRL2006.pdf
Engineering e-Infrastructures for the Benefits of e-Science Contact: D. Kranzlmueller Schedule: W29 The importance of e-Science as an ingredient of today's problem solving process in computational science and engineering is accepted as a matter of fact throughout the scientific and research community. In addition, it can be seen as a main driving factor for provisioning of e-Infrastructures, on which different tools and applications can build on for their own benefits when exploiting networking and computational resources around the globe. With the recent advances of networking and the push towards ever more sophisticated transportation mechanisms and protocols, while at the same time prototype grid infrastructures are getting mature and production ready, the basis for the researcher's daily work is steadily improving. At the same time, the dependency of the e-Science community on the functional and operational quality of the e-Infrastructures increases.
This workshop intendes to bring together all the different parties involved in the engineering of e-Infrastructures. This includes all the players involved in e-Infrastructures, from low-level networking community to high-level application developers. Consequently, topics of interest include the latest advances in networking and interconnectivity, grid middleware development and deployment, advanced computing resources and supercomputing, storage and data services, and e-Infrastructure application users and corresponding user support. It is expected that by providing this forum as a means of interaction and discussion between the different communities, an
exchange of requirements and needs, new ideas for hot issues, and especially improvements at the interfaces between the different e-Infrastructure layers may be possible.
The workshop will be organized by inviting several key representatives for the various aspects (supercomputing, grid computing, lambda networking, senso grid, ...) and to have a panel discussion on the integrational aspects of the infrastructure
and its usabillity for the e-Science based researchers. http://www.austriangrid.at/e-science
e-Humanities - an emerging area of concern Contact: P. Wittenburg Schedule: W30 In the humanities the availability of new digital technology and increasing amounts of digitized data is leading to novel empirical research methods based on computational modeling and automatic detection of patterns in unstructured data such as texts and multimedia recordings, as well as across independently compiled databases. The capability of creating and using large digital collections of structured and unstructured resources and the emergence of powerful algorithms for processing the data from multiple perspectives will deeply affect all disciplines of the humanities. A rapidly growing number of researchers in the Humanities begin to understand that the computational processing empowered by the Internet is already revolutionizing the research methods. The potential of virtually integrating the existing distributed resources (which were not designed with interoperability in mind) and virtually combining algorithms that were developed at different institutions to new workflows offers unprecedented possibilities for researchers in the humanities. Last but not least, advanced methods for structure mining in large unstructured resources require the processing and memory capacity of supercomputers. The combination of these factors (awareness of the promises, provided that interoperability issues can be solved and sufficient computer power can be made available) calls for opportunities for researchers from the humanities to meet and discuss with researchers in the field of e-Science and Grid Computing.
Currently, in the humanities domain the first Data Grid projects are being carried out to form joint digital collections that are accessible for scholars without a need to deal explicitly with interoperability problems (DAM-LR, Text-Grid). Humanities researchers are beginning to tackle some of the big challenges requiring Grid Computing (training quantitative language models, doing pattern recognition on video streams, modeling cognitive processes with the help neuronal dynamics, etc.) and some centers are in the process of creating complex workflow systems based on web services. All these activities can only be successful if the interoperability problems can be solved at all levels ranging from data encoding and syntactical aspects to the semantics of the contents and the meta-data. Standardization efforts are taken and Semantic Web technologies are being applied.
It is difficult, however, even for the specialists to oversee all the various initiatives. The workshop will give the experts the opportunity to interact, to take profit from mutual fertilization and to work out joint strategies for the future. In addition the workshop will be a forum that will help to create or improve the awareness about the emerging new eHumanities paradigm. http://www.mpi.nl/clarin/eScienceWorkshop.htm
e-Science in and Beyond the Classroom: Usability, Practicability and Sensability Contact: H.C. Smith Schedule: W32 e-Science and Grid applications have the potential to engage the wider community in scientific inquiry and debate and to transform learning by enabling active participation in distributed global collaborations through access to data, communication and computing resources. This workshop addresses the practical issues involved in bringing e-Science into the classroom, and extending its use beyond, into other formal and informal learning contexts. Making e-Science useful and usable in world-wide contexts brings its own challenges, such as widely varying IT infrastructures - the digital divide; the ethics of data access, storage and sharing; co-ordination across time-zones; differences in user expertise, expectations and culture, etc. Participants are invited to submit a maximum 4 page position paper describing their research in the area of eScience and learning. We are particularly interested in papers that address the practical issues involved in delivering eScience for learning in real contexts.
This workshop aims to provide a forum for researchers to exchange experiences and ideas about designing, developing and using e-Science applications for learning. We will explore the issues through presentations, demonstrations and discussion.
To encourage interactive discussion we propose to limit the workshop to 12 participants, and have created a pre-conference workshop blog (at http://escience06.blogspot.com) to which you are welcome to contribute in advance of the workshop. http://www.informatics.sussex.ac.uk/users/hilarys/esciencewshop/
HealthGrid Contact: Y. LEGRE Schedule: W34 This workshop is an open forum for the exchange and discussion of ideas, technologies, solutions and requirements that interest the Grid and the Life-Sciences communities to foster the integration of Grids into Health.
Participation is encouraged for Grid middleware and Grid applications developers, Biomedical and Health Informatics users and security and policy makers to participate with a common concern on the applications to Health.
Legal, economical and social aspects will be also studied and papers are expected in this field.
We expect also to receive feedback and comments on the draft roadmap for wide adoption and deployment of HealthGrid in Europe
NWO final meeting Computational Science Contact: M. de Boer Schedule: W39 This session serves as the final meeting of the NWO research programme Computational Science (2001 - 2006). Earlier on the same day, prof. Jef Huisman will give an impression of his experiences with the Computational Science programme. The title of his key-note lecture is “Complex dynamics in the plankton: computational science applied to the world's oceans”.
All researchers which are and have been working on Computational Science projects are warmly welcomed, as well as any other researcher which an interest in small scale, high quality and multidisciplinary projects. For further info on the NWO Computational Science programme please refer to www.computationalscience.nl.
The workshop will consist of three talks:
1. prof. Hester Bijl (TUD), "Simulation of fluid-structure interaction: advances and prospects"
Abstract: Interactions between fluid and structures occur in a wide range of applications. A famous example is the destruction of the Tacoma bridge by the wind. Also in aircraft, spacecraft and wind turbines unwanted fluid-structure interactions can occur. Often with desastrous effects. The design of new - more flexible - materials warrants accurate predictions of fluid-structure interactions. Due to the complexity of these simulations, efficiency is of the utmost importance. In our project we have developed numerical techniques to increase the efficiency of fluid-structure interaction simulations. These include a scheme to integrate the coupled system in time, an algorithm to transfer the information between the flouid and the solid, an algorithm to move the flow mesh, and acceleration schemes. With the resulting solver fluid-structure interaction of a wing, a typical test case, has been computed. In the future we intend to further apply our method to other fluid-structure interaction problems, especially in off-shore wind turbines. In addition we want to include uncertainties in parameters, such as inflow velocity or geometry of the blade, resulting in a stochastic simulation. Finally, we are developing a Flexible Coupling Shell which can be used by others to couple arbitrary flow and structure solvers using our techniques.
2. dr. Herman te Riele (CWI), "The complexity of factoring large numbers: theory and practice"
Abstract: The problem of factoring large numbers has attracted the interest of human computers already since Fermat, who published one of the first more sophisticated factoring algorithms. The publication, in 1978, of the RSA public-key cryptographic system, followed by its wide-spread application in digital communication, has stimulated renewed interest in the theory and practice of factoring, because the security of RSA depends on the difficulty of factoring large numbers and because the computational complexity of factoring large numbers was (and still is) unknown.
As a result, the world record for factoring large numbers has been pushed forward from about 50 digits in 1978 to 200 digits in 2005.
In this talk we will discuss the extremely CPU- and memory-intensive computations which have been carried out to achieve this progress and compare this with our (heuristic) knowledge of the complexity of factoring large numbers.
3. prof. Henk van der Vorst (Utrecht University and Catholic University Leuven), "Big Problems: fast solutions"
Abstract: The simulation of ocean flows with a computer, the calculation of the effects of cooling of infant's brains, the temperature distribution over a space rocket, trying to model the behaviour of plasma in a fusion reactor: these are only a few examples of problems that lead to the necessity to solve large systems of equations. The solution of a linear system is a tedious task and not very challenging from a mathematical point of view. However, it becomes a different matter when the system is very large, a million equations or more, say. Early in the 19th century, with no computers available, even small systems posed practical problems, which led Gauss to design a cheap method for solving his systems (with only 4 unknowns) approximately. His method is still of practical value. After Gauss, many mathematicians studied the question whether the desired solutions could be obtained faster. An important step was made by the Russian applied mathematician Krylov in the 1930s. His approach led to the now important Krylov spaces. After some early success, attention for his approach declined because it required excessive accuracy in computation and this was expensive in terms of computing time. It was Lanczos, who in 1952 proposed a variant of the Krylov approach that became extremely successful. This was not the end of the journey, because Lanczos' method could only be used for special problems and other challenging problems waited for their solution. I will try to sketch a picture of the research in this very active area and indicate the fabulous factors saved in computing time.
Finally, the concluding publication of the Computational Science programme will be presented. For further info: http://www.computationalscience.nl/news/