Keywords: high-throughput computing, cloud computing, Web 2.0, performance-based earthquake engineering, seismic assessment of structures, ICE4RISK.

Existing high-performance and high-throughput computing facilities, such as computing clusters, big data-centres, service infrastructure and test beds, are currently restricted to elite stake-holders familiar with distributed service infrastructures, programming paradigms, workflow composition systems and their respective security infrastructure. There is, an emergent need for a novel distributed software platform capable of addressing the real-world needs of the scientific and engineering communities and industries by allowing them to exploit distributed resources and solve complex problems through more secure, intuitive and technology-hiding means leveraging both the current networking and advanced computing environments.

It is reported that the risk to be killed due to earthquakes in Europe in the last twenty years is about eight times higher than that observed in Japan and even eighty times higher than in the United States. High observed losses point to the need for improved seismic design and risk assessment approaches capable of achieving more accurate and predictable results. Since the seismic risk assessment is computationally extremely demanding and consequently time-consuming, making it difficult for practical application, it is necessary to develop a user-friendly and high-throughput computational environment for seismic risk assessment. The innovative methods for seismic risk assessment based on the nonlinear static or dynamic analyses and are usually probabilistic and not determinist, which is a regular practice in design offices. In order to estimate the seismic risk of selected structures many nonlinear simulations have to be performed. It is therefore almost impossible to perform probabilistic seismic risk assessment for practical purpose, since the user friendly tools are not available.

To provide the above features the paper examines different existing and emergent technologies, including: Web 2.0 [1] and cloud computing [2]. The described end-user scenario focus on the use of parametric studies in earthquake engineering. The scenario paints a vision of a future computing environment by describing a probabilistic performance assessment of a structure [3] as an example for a high-throughput computing environment. The adopted system architecture was selected primarily because it provides a scalable computing solution without the administrative overhead of a local computing cluster, and because it allows users to carry out most of their science with only a web browser.

The paper provides an overview of the ICE4RISK project [4] as well as brief descriptions and discussion of the underlying technologies used in the development of the developed computing environment. A basic overview of the ICE4RISK computing environment architecture is also presented.

1

T. O'Reilly, "Web 2.0 Compact Definition: Trying Again", O'Reilly Radar, December 10, 2006. URL

2

M. Miller, "Cloud Computing: Web-Based Applications That Change the Way You Work and Collaborate Online", Que, 2008.

3

M. Dolenc, M. Dolšek, "High-Throughput Computing in Engineering", in M. Papadrakakis, B.H.V. Topping, (Editors), "Proceedings of the Sixth International Conference on Engineering Computational Technology", Civil-Comp Press, Stirlingshire, UK, Paper 94, 2008. doi:10.4203/ccp.89.94

4

ICE4RISK, "High-throughput computing environment for seismic risk assessment". http://ice4risk.slo-projekt.info

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £125 +P&P)