Seminars

Academic Year 2012 - 2013 Seminars

Dan Katz, Program Director, National Science Foundation
Title:  "Exploring Application Paradigms with Montage"
Date: Friday November 9nd, 2pm
Location: CoRE 701

Abstract:

This talk will use Montage, an astronomical image mosaicking application that is a toolbox of independent components, to explore various application paradigms on parallel and distributed systems, as the Montage components can be used in a variety of settings, including on a single system, on a parallel system, or on a set of distributed systems, including grids and clouds. Montage, which was built to use MPI in parallel, and Pegasus/DAGman on distributed systems, has also been used as an exemplar many-task computing (or workflow) application by a number of other tool and system developers. In this talk, a variety of work with Montage will be discussed, including the use of multiple types of infrastructure/middleware, the use of scripting to allow a user to easily customize their use of the Montage components, and overcoming data management issues.

About Dan Katz:

Daniel S. Katz is a Program Director in the software cluster in the Office of Cyberinfrastructure at NSF, and is also a Senior Fellow in the Computation Institute (CI) at the University of Chicago and Argonne National Laboratory, an affiliate faculty member at the Center for Computation and Technology (CCT), Louisiana State University (LSU), and Adjunct Associate Professor in the Department of Electrical and Computer Engineering at LSU. His NSF interests include issues related to sustainable software, such as open source community governance, metrics, attribution, and reproducibility. His research interests include: numerical methods, algorithms, and programming applied to supercomputing, parallel computing, cluster computing, distributed computing, and embedded computing; and fault-tolerant computing. 

 Matteo Turilli,   Senior Research Associate, Oxford e-Research Centre
Title:  "The EGI Pan-European Cloud Test Bed: Use Cases and Federation Model"
Date: October 16th, 2012, 2:00 - 3:30pm
Location: CoRE 701

Abstract:

A growing number of research communities are looking at cloud computing and its underlying economic and technological models as a way to obtain cheap, scalable and easy to access computational resources. Several cloud initiatives have cropped up across Europe following this demand, both in the private and public sector. The talk presents the cloud activities of the European Grid Initiative (EGI) and focuses on the model of federation that has been chosen to deploy an EGI pan-European cloud test bed. Several use cases are reviewed to show how user communities intend to benefit from cloud computing and what kinds of capabilities a federation of clouds should offer. Different models of federation are introduced and a detailed analysis of the model chosen by EGI is offered. The implementation of the EGI federation test bed is discussed with attention to i) cloud standards, brokering and clients; ii) how cloud resources are integrated within a pre-existing infrastructure; and iii) how they compare to other ways to share and distribute computing and data resources to the research communities.

About Matteo Turilli:

Matteo Turilli is a Senior Research Associate at the Oxford e-Research Centre. He holds a DPhil (PhD) in Computer Science from the University of Oxford and is involved in several projects concerning the research and development of Cloud Computing. Matteo's main research interest is in parallel and distributed computing, specifically as it relates to ethical and security requirements. Currently, Matteo is the leader of the EGI-InSPIRE Task on Federated Clouds and member of the Oxford Cyber Security Centre and of the Oxford Requirement and Foundations Group.

Neil Chue Hong, Director, Software Sustainability Institute
Title:  "Doing Science Properly in the Digital Age"
Date:  Monday, October 1st from 3:00 - 4:30pm
Location:  CoRE 701

Abstract:

Science has changed dramatically over the last seventy years. The rise of computational simulation and data-intensive research – the third and fourth paradigms of science – to take their place alongside the established theory and experimental methods, has ushered in a new digital age of science. This has led to pervasive use of software across most scientific disciplines, ranging from monolithic codes to web-based execution environments.

With this increase in computation and data comes associated challenges. How do we deal with the increasing amounts of data? As the size and geographical distribution of collaborations increases, how do we ensure that the network does not become more fragile? Can we record all the metadata associated with each step of our research workflow? Most importantly, can we convince our peers that the outcomes of our research are useful?

In my work with the Software Sustainability Institute, I have had the opportunity to collaborate with both researchers developing and using software, as well as researchers studying those who develop and use software. This talk examines some of the issues and approaches to dealing with scientific software as a ubiquitous part of research.

The challenge facing the new generations of researchers is clear: what does it mean to say that we are doing science properly in the digital age?

Sandra Gesing, Senior consultant and PhD Student in the Applied Bioinformatics Group, The University of Tübingen
Title:  "Web-Based Science Gateways for Structural Bioinformatics"
Date:  Friday, September 28th from 3:00 - 4:30pm 
Location:  CoRE 701

Abstract:

Structural bioinformatics is concerned with computational methods to gain new insight into molecular structures, their prediction, and the analysis of their functions. Nowadays, these computational methods are invaluable tools in numerous applications like materials science, structural biology, and drug design. DCIs (Distributed Computing Infrastructures) allow gathering information about increasingly complex chemical structures via available sophisticated tools and methods. Web-based science gateways support users to access tools for a specific application domain via a web browser. The overall goal for creating science gateways is to increase the usability of tools and to hide the complexity of the underlying infrastructure. In the project MoSGrid (Molecular Simulation Grid) a complete solution for the structural bioinformatics community has been developed which aids end users and developers to use tools and workflows in DCIs.

The talk goes into detail for the role-based user management concerted for the structural bioinformatics community, the credential management facilitating SAML (Security Assertion Markup Language), a tool for job and workflow management allowing workflow interoperability, and an extended framework for the migration of workflows supporting the re-usability of workflows for different DCIs.