In order to achieve the objectives of the application, it will be fundamental to incorporate the results obtained in the areas of digital content, digital repositories and remote laboratories.
Digital content: As part of the grid infrastructure, state-of-the art multimedia servers will be used. These servers employ advance techniques to retrieve and transmit interoperable and multiplatform information (videos, slides, and other media) in order make efficient use of the available resources. They are totally scalable, based on free software, and they can be adapted to different environments with different access bandwidths. Transparencies can be stored synchronized with video material. These transparencies are programmable in such a way that the student can interact in real time with them and have access to other class material such as simulators, etc.
Digital repository: Distributed contents and databases represent a solution for sharing content. Developed under international standards they also help to make easier and personalised searches. Learning objects and multimedia repositories, already developed, represent a model to be used for semantic grid.
Remote laboratory: Remote laboratories are specific activities that can be shared among participants from different places, with different methodologies and infrastructures, where interdisciplinary scientific courses can be given through the network, taking advantage of the bandwidth for high computing distributed projects. A special and very important kind of remote laboratory is the one where geographically distributed people can simultaneously access and control in real time an experimental set up. People performing remotely an experiment in a collaborative way need to access distributed contents and databases. Recent advances in GRID computing and computer-controlled instrumentation permit net-based techniques to be utilized for setting up remote laboratory access.
These considerations justify the use of GRID based new tools in e-Education. To begin with, a remote laboratory for the food engineer career will be considered (Laboratorio Experimental Multidisciplinario a Distancia – Proyecto EN 213904 – PAPIME – Mexico). In parallel, an experimental set-up suitable for quantitative study of mechanical oscillations (currently in implementation phase) will be held in Rio de Janeiro (CECIERJ/CEDERJ and UFRJ) and accessed remotely by EELA partners (initially, CECIERJ/CEDERJ, CIEMAT, CUBAENERGIA, UFRJ and UNAM).
If e-education is to help shorten the digital divide, it is reasonable to suppose that many individuals involved in the process are not familiar with advanced technology. Therefore, it would be quite difficult to issue complex experiments at a first moment. In the context of EELA, remote labs can be seen as a means of introducing both students and teachers into the GRID culture while laying the basis for future development of computational demanding experiments (e.g., eSTAR project). In spite of the fact that Remote laboratory will not immediately take advantage of the Grid technology, its integration with Digital content and Digital repository applications (already gridified) can provide a paramount foundation for the new era e-Education.
Theses experiences will help to:
- Provide teachers and tutors at Universities with tools to access the grid in order to prepare their courses, enhancing the distributed collaborative work;
- Help students to learn about grid technology by using it to solve problems which are computer intensive (computer intensive problems should be understood as number crunching problems as well as complex symbolic processes) and accessing remote laboratories using communication and control technology;
- Adapt content under international standards in order to have distributed digital libraries and repositories;
- Create courses to be given in a collaborative way.
The following possible activities are foreseen:
- Remote laboratory
· Remote laboratory customization requirement analysis
· Customization of the Grid middleware chosen by EELA
· Design of the remote laboratory web interface suitable for EELA activities
· Remote laboratory web interface certification
· Remote laboratory use in courses and dissemination
· Impact on educational processes
- Multimedia Interactive Course
· Repository and software analysis and interactive course identification
· Set up of multimedia data repository
· Set up of multimedia grid virtual server
· Multimedia repository and server certification
· Update of multimedia interactive material
- Semantic Grid
· Semantic requirements capture
· Customization of the Grid middleware chosen by EELA
· Semantic Grid middleware certification
· Customization of the semantic Grid for dissemination tools
· Semantic Grid use in courses and dissemination
· Impact on educational processes
- Collaborative Courses
· Set up tutor framework
· Design of collaborative courses
· Pilot course deployment
· Monitoring of the pilot course
· Implementation in regular science courses
· Impact on educational processes
The following activities are currently in progress and should be ported on the EELA infrastructure in 2007:
- CuGfL
Cuba Grid for Learning (CuGfL). CuGfL is a One-Stop-Center for quality assured online learning content with the aim to promote and support the lifelong learning agenda in Cuba to accelerate the growth of K-Society. The main objectives of CuGfL are:
- To enhance discoverability of e-learning content form heterogeneous sources;
- To develop e-learning standards to ensure conformance and adoption of best practices in e-learning content and systems;
- To provide e-Learning systems and tools to enable and support e-Learning activities and processes for the purpose of life-long learning;
- To encourage sharing and development of local/indigenous content.
This way, CuGfL appears as a general framework to guide the construction of other Learning Management systems. CuGfL has a portal located at http://www.cursosenlinea.cu and http://www.redciencia.cu. In order to incorporate the results obtained in the areas of digital content, digital repositories and remote laboratories this group is mainly interested in working on the Learning Management System (LMS), digital content, multimedia repository and e-learning services in order to:
- Test and adapt the latest version of LMS, e-learning services, methodologies, infrastructures and multimedia repository, on grid and IPv6 environment with different access bandwidths.
- Customize and optimize the LMS and e-learning service redesigning the applications and service management according to the grid and IPv6 environment. Upgrade the e-learning tools, simulators, virtual laboratories, digital repository and multimedia servers mainly based on free software.
Theses experiences will help to:
- Provide teachers and tutors at scientific institutes and Universities with tools to access the grid in order to prepare their courses, enhancing the distributed collaborative work;
- Help students to learn about grid technology by using it to solve problems and accessing remote laboratories;
- Adapt content under international standards according the distributed digital libraries and repositories;
According to their resources, they are interested in taking part on the following activities:
- Remote laboratory
- Multimedia Interactive Course
- Semantic Grid
- Collaborative Courses
- VoD
This e-learning application consists of a distributed interactive multimedia server (RIO) that is currently being employed in the distance learning consortium of public universities in the state of Rio de Janeiro (CEDERJ). Presently, students of the Computer Systems 3-year graduate course access the RIO video server to stream lectures of each class. The video stream is coupled with class slides and a brief index of the topics in the lecture. Students can interact with the slides, which are programmable entities and can launch other applications in the client&rdquot;s computer. The RIO server is made of storage units (storage servers) that may reside in different machines at different locations. For the EELA project the RIO server will be adapted to run over the GRID paradigm. The goal is to provide an underlying structure that allows dynamic reconfiguration of resources. As part of the E-learning initiative for this task, a prototype virtual lab is being built. The prototype will be tied to the multimedia server clients&rdquot; environment. This way, students watching a video course would be able to interact with the class slides and, from those, access the prototype virtual lab while guided by the video-lecturer. More information about the VoD-RemoteLab, available software and videos can be found at http://trindade.land.ufrj.br/~vod/rio/.
- LEMDist
LEMDist is not only an application, but a project itself. Thus, its aim is to get web access to laboratory equipment and another web service to help e-science and e-learning users.
The project is driven to build technological support for different pedagogical approach in natural science teaching, and heterogeneous problem solving environment for scientific work. The system is based on applications with access to distributed computer enhanced instrumentation, remote access to simulation capabilities with high performance computing support, interactive visualization, distributed data analysis and access to heterogeneous data sources systems.
For doing this, five layers for the architecture have been identified:
- Access layer: A web user interface.
- Service layer.
- Grid layer
- Administration and security layer.
- Resource layer.
The actual goal is to allow remote access to about four tenths of laboratory equipment and web service based on the gLite environment. The system implementation is based on message passing and procedure execution tasks.
The first steps were made with the Multidisciplinary Experimental Laboratory (LEM) equipment, which has different kinds of instruments for measurement and processing in several fields: Unitary operations, Industrial process simulation, Control process, Physical, Chemical and Physical Chemistry properties determination, etc.
The research in the remote access will be focused in several devices. The most important will be a 200 Mhz Nuclear Magnetic Resonance (NMR) Varian spectrometer, a Capillarity Electrophoresis (CE) equipment and the associated computer controlled laboratory apparatus through a serial interface.
The plans for these next months are:
- Grid environment: Build a training Grid with gLite middleware.
- Port the NMR access to the Grid environment.
- Design an NMR short course.
- Select a software for NMR simulation (cns_solve is a good candidate for computational modeling).
- Develop visualization objects for NMR.
- Develop visualization objects for Microwave experiments.
- Develop the remote access interface for the capillarity electrophoresis system with LABView.
- Build the web user interface.
Other fields of work to be tackled in the future will be chemical and food engineering and Industrial chemistry.
- PILP
This application is intended to discover hidden data from relational databases. It uses a technique called Inductive Logic Programming (ILP), where, given a background knowledge, a set of positive examples, a set of negative examples, and a language bias, the objective is to generate first order rules that (almost) perfectly describes all positive examples and none of the negative examples. We have been working with several domains, when applying ILP: drug discovery, analysis of mammograms, link discovery, among others. These domains present very large databases and sets of examples. This application has already been executed in a grid environment.
- SATyrus
SATyrus is a novel approach to the specification and solving of optimization problems. In the scope of the SATyrus architecture, a given target problem is specified, using a logical style declarative language, as a set of pseudo-Boolean constraints. Then, this set of constraints is compiled, following a satisfiability (SAT-based) mapping, into an energy function representing the space state of solutions of the target problem. Among other possible computational intelligence models that could have been adopted (e.g., genetic algorithms, artificial immune systems, etc), higher-order Hopfield networks of stochastic neurons were chosen to map and minimize the resulting energy function. This choice is justified by the linear cost of representing any energy function produced by SATyrus as symmetric neural networks.
Advantages of using logical constraints as a description language for optimization problems are discussed in this paper. Finding an IT model for a logical sentence is a problem that apparently does not involve optimizing a cost function. In propositional logic that corresponds to the assertion of truth-values to the propositional symbols that appear in the formula in question, in such a way that the formula as a whole becomes true. Nevertheless, it is possible to construct a function that maps the truth-value of a formula into the set {0,1}. Some problems may be better described as a combination of logical and mathematical constraints. A subset of this combination could be seen as a sum of weighted products of boolean variables, pseudo-Boolean constraints.
|