The International Collaboration to Extend and Advance Grid Education

Description:

Aims

E-science is a major priority in international research and development. This focus of effort in both University research and industrial R&D is driving the increasing demand for individuals trained in E-Science technology and methodology. This trend is supported by the development of Grid technologies and the convergence of the Grid and Web Services to form the Semantic Grid. The training provided by this EPSRC-funded degree programme builds upon core computing knowledge developed through a BSc programme to develop advanced knowledge in distributed computing and the Semantic Grid. Furthermore, E-Science operates as a technology driver for other application areas including E-Commerce, E-Health and E-Governance.

Requirements

The course is aimed at graduates with a 2(ii) or better degree in computing science or a related discipline. Graduates in other disciplines may be admitted providing that they have an appropriate level of qualification in IT.

Syllabus

The Electronic Society, Technologies for the World Wide Web, Enterprise Programming, Security & Privacy, E-Science and Grid Computing, Multi-Agent Systems, Data Interpretation and Communication, E-Science Workshop.

For the remainder of the 12 month MSc programme, students undertake the MSc Project in E-Science Technology. These projects are informed by current research and development in E-Science, and may be jointly supervised by a member of academic staff and an external collaborator.

Description

Starting from September 2005 we have defined 5 new programmes within the Master Grid Computing:

1. Computational Science

Computational Science focuses on systems for quantitative modelling and simulation of complex dynamic systems. These are found in, e.g., physics, chemistry and engineering. Key words: mesoscopic modelling, cellular automata, parallel algorithms, scientific visualisation and virtual reality.

2. Bioinformatics

Bioinformatics focuses on modelling and analysis of biological systems. Key words: analysis of molecular biology databases, sequence analysis, modelling of regulatory networks and metabolic pathways, systems biology, biomedical applications.

3. Computer Systems Architecture

Computer Systems Architecture focuses on parallel and distributed hardware and software architecture for high performance computing. Key words: Grid systems, resource management, high-speed networks , next-generation computer architectures, systems on a chip.

4. Program algebra & generic programming

Program algebra & generic programming focuses on programming environments based on formal language definitions, process theory and tools for the specification and verification of concurrent communicating systems. Key words: code re-factoring, software renovation, intermediate languages, process algebra and executable specifications.

5. Free Programme in Computer Science

After consultation with the Master Team Grid Computing, you have the possibility to compose your own programme in computer science.

Each programme consists of 5 mandatory topics

• Introduction to Grid Computing
• Distributed Programming Methods / Introduction to Bioinformatics (bioinformatics programme)
• Distributed Stochastic Simulation
• Grid Hardware Infrastructure / Advanced Bioinformatics (bioinformatics programme)
• Profile project

In semester II, year 1 + semester I, year 2

• 4 Programme-related courses
• 3 Free choice
• Graduation project

Description:

This new MSc programme is planned to start in September 2006. It is proposed that the following modules will make up the course, although this is subject to change as planning and development of the course continues. Please contact the CEMS Graduate School if you have any queries in this regard.

• Grid: Virtual Organisations
• Algorithms
• Advanced Databases
• Distributed Systems
• Grids and Application Domains
• Communication Networks and Protocols
• Intelligent Adaptive Systems
• Security in Networks
• Semantic Grid
• Research Methods
• Computer Science Dissertation

Description

The programme aims to provide an advanced theoretical background to Grid computing for building large-scale distributed systems. It will cover advanced theory of distributed systems with a focus on Grid computing. It will also develop a detailed technical knowledge of current practices in Grid computing applications. It is a distinctive programme that is not taught anywhere in the UK in terms of the combination of distributed systems and Grid computing. The proposed programme will not only offer advanced theory but also the practical aspects with the support of a fully working computing laboratory in the School which enables students to develop their skills in this field.The aim of the programme is to equip high quality and ambitious graduates with the necessary advanced technical and professional skills for an enhanced career either in industry or leading edge research in the area of distributed systems and Grid computing. Specifically, the main objectives of the programme are:

• To critically appraise advanced technologies for developing Grid systems;
• To practically examine the development of large scale Grid systems;
• To critically investigate the problems and pitfalls of Grid systems in business, commerce, and industry.

Modules

• Advanced Distributed Systems
• Computer Networks
• Network Computing
• Network Security and Data Encryption
• Grid Middleware Technologies
• Grid System Analysis and Design
• Project management
• Workshop
• Dissertation

This one year Grid Computing course consists of the following modules:

• Grid Computing Fundamentals
• Grid Middleware (Linux/Globus at source code level)
• Grid Infrastructure (Distributed Systems, Networking, Communication)
• Grid Development and e-Engineering Applications (Group Project)
• High Performance Scientific Computing on the Grid (Parallel Computing, Data Mining, CAD)
• Management for Technology
• Visualization
• Software Engineering
• C++ Programming
• Java Programming

The MSc Programme

Edinburgh is building on its prominence in e-Science to offer an exciting one-year postgraduate programme . Students can either take two semesters of taught courses leading to a Postgraduate Diploma in e-Science, or supplement that with an extended individual research project for an MSc degree in e-Science. The programme provides students with an appreciation of the principles underlying e-Science and with practical experience of using current e-Science technologies. This expertise, together with the transferable skills developed during the programme, will provide an equally good preparation for students wanting a career in scientific research and in the commercial IT sector; e.g. in the biomedical or pharmaceutical industries, where similar challenges exist.

The core of the programme is the following set of mandatory courses:

• Distributed Computing for e-Science 1
• Software Engineering with Objects and Components
• Introduction to Scientific Data
• Programming for e-Science
• Distributed Computing for e-Science 2
• Software Architecture, Process and Management
• Topics in e-Science
• Project Preparation (e-Science)

Students also take four optional courses, drawn from a large pool, which enables them to tailor their personal curriculum to their individual interests. Possible options include courses from Informatics, Physics and Astrophysics, Geographical Information Science and additional courses covering the application of e-Science to particular scientific domains, such as `The Virtual Observatory' for astronomers.

EPCC, the Edinburgh Parallel Computing Centre, offers a one-year taught Masters course: the MSc in High Performance Computing (HPC). This well-established programme provides an excellent grounding in HPC technologies and their practical application.

The use of powerful computers and networks is becoming ever more important in science, technology and commerce. Many new and exciting research areas are opening up, for example in physics, chemistry, engineering and biology. Studying these topics using HPC requires leading-edge knowledge of applied computer science.

The MSc in HPC will appeal to students who have a keen interest in programming and would like to learn about HPC and parallel computing. We aim to attract students who hold degrees in the physical sciences, engineering, computer science or mathematics, or who have equivalent work experience.

The course has a strong practical focus and students have access to world-leading HPC platforms and technologies. These include HPCx, the 1600-processor UK national academic supercomputer service, and BlueSky, a leading-edge IBM BlueGene system operated by EPCC.

The MSc is awarded by the University of Edinburgh. Graduates will hold one of the few university-accredited postgraduate HPC qualifications in Europe.

A number of funded studentships are available for UK and EU residents. It is also possible to study part-time: please contact the MSc programme coordinator for more information. We also offer, in our capacity as an HEC Graduate Training Centre, the opportunity to combine our Masters course with a PhD from another UK institution.

EPCC is an institute within the School of Physics at the University of Edinburgh. It has an international reputation as a world-leading centre of expertise in HPC and Grid computing. The MSc is supported by EPSRC.

Aims

To provide the participants with:

• Detailed understanding of the key problems and issues that arise when attempting large-scale distributed computation, both within organisations and across organisational boundaries
• Insight into the architectural implications of Grid-scale computation
• Awareness of current research issues in:
• Grid architecture and infrastructure
• Scalable distributed computation
• Integration of applications across autonomous organisations
• Practical experience of current Grid technologies and the associated standards
• Skills in utilising current Grid tools and technologies
• Appreciation of the weaknesses of existing tools and technologies, and potential areas for improvement

Objectives

By the end of this module, participants should be able to:

• Critically discuss and reason about large-scale distributed system architectures, infrastructures and technologies
• Articulate research challenges in multi-organisational distributed computing, including Grid computing
• Design and implement Grid computing applications using Globus or similar toolkits
• Justify the applicability, or non-applicability, of Grid technologies for a specific application

Module Content

The module consists of 20 lectures and 10 tutorial & laboratory sessions. Additional time will be required for reading, and at least 30 hours of unscheduled practical work will be expected. The detailed content listed here is indicative of the likely course content rather than being an exact prescription.

• The first two weeks of the module will consist of six lectures, providing a review of general distributed systems issues, web services and XML, networking and communications technology, and an introduction to large-scale systems architecture.
• In weeks 3 to 10 of the module there will be a weekly tutorial on current Grid-related technologies, along with associated reading and practical work, focussed on system design and implementation using Globus or a similar technology.
• In parallel with the practical activities, there will be a further 14 lectures covering:
• systems architectures;
• scheduling, accounting and resource management;
• scalability;
• systems modelling and simulation; and
• security

...in the specific settings of large-scale multi-organisational distributed computing and large-scale cluster computing.

Course structure

There are six streams, each of which covers an area of specialisation. Students are normally required to select one of these areas of specialisation (below), although students with relevant previous experience or special interests may be allowed to take different combinations of course options with the agreement of the course director.

There are two different pathways through the course: the standard pathway is for students primarily interested in a career oriented towards development and applications in industry; the research pathway is for students interested in a research career, either in the industrial or academic sector.

Computational management

This stream offers a course of study on the theory and tools of business management that require computerised solutions. Graduates will be well-equipped to contribute to academic research and to commercial, industrial and financial applications.

Logic and artificial intelligence

This covers the theory and application of artificial intelligence, with emphasis on the use of logic as a unifying framework. Options within this area allow students to specialise in the design and implementation of applications, or in the use of various classical and non-classical systems of logic in the resolution of fundamental questions in artificial intelligence and problems in computer science generally.

Mathematical foundations

This focuses on the applications of mathematics to the theory and practice of computing. On completion of the course, a specialist will be able to make contributions to current research into finding new methods needed to strengthen a systematic approach to developments in computing. Software engineering This is intended for computer science graduates who are looking for an advanced course of study in the methods, tools, techniques and processes underlying the development of large and complex software systems. Particular emphasis is placed on solving problems due to software size and age. Graduates will be well-equipped to contribute to both academic research and industrial applications. High performance computing This area of specialisation offers a course of study in techniques for solving computationally intensive problems in acceptable time. Applications of this type are becoming increasingly important in many areas of science, engineering, medicine, finance and industry. Particular emphasis is placed on combining a knowledge of architectures, especially parallel architectures, algorithms and systems software to meet performance goals. Graduates will be well-equipped to contribute to academic or industrial research in the underlying technologies, or to their practical application. Parallel and distributed systems Intended for computer science graduates who are looking for a course that focuses on the tools, techniques and technologies of concurrent systems, this area of specialisation places particular emphasis on the design of distributed and parallel algorithms. Graduates will be well-equipped to contribute both to academic research and industrial applications.

The course modules

(some of which may not be offered every year) are as follows:

• Advanced computer architecture
• Advanced databases
• Advanced issues in object-oriented programming
• Advanced operational research
• Advanced graphics and visualisation
• Advances in artificial intelligence
• Complexity
• Computer vision
• Computing for optimal decisions
• Custom computing
• Grid computing
• Intelligent data analysis and probabilistic inference
• Knowledge representation
• Modal and temporal logic
• Models of concurrent computation
• Multi-agent systems
• Natural language processing
• Network security
• Parallel algorithms
• Performance analysis
• Program analysis
• Quantum computing
• Software engineering environments
• Type systems for programming languages

Aims

This course unit aims to:

1. explain the concept of eScience and its importance in future problem solving IT infrastructure.
2. explain the concept of Grid computing and its relation to eScience,
3. familarise students with the key abstractions underpinning the Grid concept,
4. outline current Grid solutions and how they are intended to evolve,
5. give a more in-depth view of a widely used Grid middleware system UNICORE,
6. give lab sessions in running Grid computing jobs using the UNICORE GUI based job composition and submission method,
7. provide a mini-project to explore some particular aspects of Grid computing, e.g. resource discovery, application plugins, workflow composition.

Learning Outcomes

A student successfully completing this course unit should:

1. Have an understanding of the concepts of Grid computing and eScience and why they have assumed such current prominence. In particular to have an understanding of the importance of standards and protocols in Grid computing (A),
2. Understand the architecuture of the UNICORE middleware and how this relates to the emerging Open Grid Services Architecture proposals and standards (A,B)
3. Be able to utilise UNICORE to submit both simple and multistage computing jobs onto a local Grid (A,B,C),
4. Be able to explore via UNICORE a particular aspect of Grid computing, for example in obtaining information about resources on wide area Grids, extending the UNICORE system via an application specific plugin, investigation interoperability with other Grid systems (e.g. Globus) (A,C).

Assessment of learning outcomes

Learning outcomes (1) and (2) are assessed by examination, learning outcome (3) is assessed by laboratory reports, learning outcome (4) is assessed by mini-project.

Contribution to programme learning

A1, A2, B2, B3, C1, C3, D1, D5.

Detailed syllabus

First we define the lecture syllabus. Each lecture will be of one hour duration. Each topic will have 3 lectures devoted to it making 12 lectures in all. The rest of the time will be devoted to the laboratory classes listed below.

1. The metacomputing problem: forerunner to the Grid. Exploring the convergence of exploitation of high speed networks, exploitation of architectural affinity, work on coupled multiphysics problems, e.g. Climate Models importance of locality requirements to minimise flow of data across wide area networks.
2. Grid computing: a persistent metacomputing environment. Digital certificates as a persistent and scalable form of authorisation, Virtualisation of resources, hiding of complexity of metacomputing environment from user.
3. Role of middleware in Grid computing. Neccesity for abstractions in a heterogeneous environment, differing OS's, resource management systems, programming languages. Interoperability achieved via tiered middleware architectures.
4. Abstract modelling approach to middleware problem - UNICORE Concept of an Abstract Job Object and its relation to workflow composition and enactment. Concept of Incarnation from abstract resource space to concrete resource space. Vertical integration in UNICORE, difference between a tiered and a layered model.

The laboratory sessions will cover the rest of the time:

• Use of UNICORE GUI to compose a Grid workflow
• Use of UNICORE Resource Broker to locale a suitable machine or machines on a local Grid
• Submission and monitoring of the job via the UNICORE client.
• Dealing with job termination and tidy up.
• Architecture extension: installing a simple client plugin for UNICORE

The key aspects of this Masters program are:

• It has been designed to provide expertise for developing service-oriented, enterprise computing systems and applications that need to operate in wired/wireless network environments. These enterprise systems can scale from a single to multiple organisations.
• It promotes the utilisation of industry standard distributed computing technologies such as J2EE and .NET.
• About half of the course focuses on highly specialised distributed computing topics such as distributed systems, cluster and grid computing, distributed algorithms, mobile systems programming, sensor networks, and Web services.
• A compulsory team-based project work that emphasises the design and development of distributed computing technologies and their application in e-Science or e-Business areas.

Objectives

Upon completion, a graduate of the program should:

• have substantial expertise in key areas of Internet programming and distributed computing,
• be able to apply acquired techniques and knowledge to contribute to the development and implementation of enterprise software systems in organisations,
• be able to analyse and design ICT projects and future ICT needs, and
• be able to apply Internet-based distributed computing systems and algorithms to e-Science and e-Business applications.

Curriculum

The MSc provides you with theory, technology and training in three complementary computer science fields that together will make you an expert in Grid Computing: Computing focuses on systems for quantitative modelling and simulation of complex dynamic systems. These are found in e.g. physics, biology and engineering. Keywords: mesoscopic modelling, cellular automata, parallel algorithms, virtual laboratories, scientific visualisation and virtual reality. System Architecture focuses on parallel and distributed Hardware and software architectures as integrated systems for high performance scientific computing and data processing. Keywords: Grid systems, resource management, high-speed networking, middle-ware technology, next-generation computer architectures, systems on a chip, system-level embedded systems design. Programmming focuses on programmming environments based on formal language definitions, process theory and tools for the specification and verification of concurrent communicating systems. Keywords: code re-factoring, software renovation, intermediate languages, linear projective semantics, hybrid process algebra and executable specifications.

You start with courses m Computing, System Architecture and Programming, later on you can select elective courses. A part of these can be taken from related courses at the participating universities. During the second year you do research and work on you r master's thesis together with experts from associated laboratories and industry.

Research Master programme

The programme is a research master's programme. Apart from basic contents, the courses will train you in the academic style and in research work.

Course Description

This is a new course, for entry in September 2007

Grid computing is an emerging computing paradigm. It provides the ability to perform high throughput computation by taking advantage of many computers connected through Internet creating a virtual architecture. This is able to distribute process execution across a parallel infrastructure. Grids perform computation on large data sets, by breaking them down into many smaller ones, or execute many more computations than would be possible on a single computer. The course will provide a comprehensive and systematic understanding of the issues, technologies and processes applicable to cluster and Grid computing. It will enable the selection and application of appropriate tools, techniques and methodologies for the successful design and implementation of complex and novel cluster and Grid computing applications and systems. The course will foster critical evaluation of academic research and industrial practice.