NEKTAR, SPICE and Vortonics: Using federated grids for large scale scientific applications (original) (raw)
Related papers
1Nektar, SPICE and Vortonics: Using Federated Grids for Large Scale Scientific Applications
2015
— A joint call for proposals was issued in late 2004- early 2005 by the US’s NSF and UK’s EPSRC for applications that aim to utilize the combined computational resources of the US and UK. In response to the call, three computational science groups from UCL, Tufts and Brown Universities teamed up with a middleware team from NIU/Argonne to meet the challenge. Successful demonstra-tions of these applications at SC05 established that not only can applications make effective use of resources that span computational grids but also underscored the need to run at the even larger scales of grids-of-grids. Although the groups had three distinct codes and aims, the projects had the underlying common feature that they were comprised of large-scale distributed applications which required high-end networking and advanced middleware in
2002
The growth of the Internet and the advent of the computational Grid have made it possible to develop and deploy advanced computational collaboratories. These systems build on high-end computational resources, communication technologies and enabling services underlying the Grid, and provide seamless and collaborative access to resources, applications and data. Combining these focused collaboratories and allowing them to interoperate has many advantages and can lead to truly collaborative, multidisciplinary and multi-institutional problem solving. However, integrating these collaboratories presents significant challenges, as each of these collaboratories has a unique architecture and implementation, and builds on different enabling technologies. This paper investigates the issues involved in integrating collaboratories operating on the Grid. It then presents the design and implementation of a prototype middleware substrate to enable a peer-to-peer integration of and global access to multiple, geographically distributed instances of the DISCOVER computational collaboratory. An experimental evaluation of the middleware substrate is presented.
The Computing and Data Grid Approach: Infrastructure for Distributed Science Applications
Computing and Informatics / Computers and Artificial Intelligence - CAI, 2002
Grid technology has evolved over the past several years to provide the services and infrastructure needed for building “virtual” systems and organizations. With this Grid based infrastructure that provides for using and managing widely distributed computing and data resources in the science environment, there is now an opportunity to provide a standard, large-scale, computing, data, instrument, and collaboration environment for science that spans many different projects, institutions, and countries. We argue that Grid technology provides an excellent basis for the creation of the integrated environments,that can combine the resources needed to support the large-scale science projects located at multiple laboratories and universities. We also present some science case studies that indicate that a paradigm shift in the process of science will
Engineering an interoperable computational collaboratory on the Grid
Concurrency and Computation: Practice and Experience, 2002
The growth of the Internet and the advent of the computational Grid have made it possible to develop and deploy advanced computational collaboratories. These systems build on high-end computational resources and communication technologies underlying the Grid, and provide seamless and collaborative access to resources, services or applications. Combining these focused collaboratories and allowing them to interoperate has many advantages and can lead to truly collaborative, multi-disciplinary and multi-institutional problem solving. However, integrating these collaboratories presents significant challenges, as each of these collaboratories has a unique architecture and implementation, and builds on different enabling technologies. This paper investigates the requirements and an architecture for interoperability among collaboratories on the Grid. It then presents the design of a middleware substrate that addresses interoperability, and a prototype implementation of this middleware substrate, to enable a peer-to-peer integration of and global collaborative access to multiple, geographically distributed instances of the DISCOVER computational collaboratory. An experimental evaluation of the middleware substrate is also presented.
Grid Computing: The Next Decade -- Report and Summary
arXiv (Cornell University), 2012
The evolution of the global scientific cyberinfrastructure (CI) has, over the last 10+ years, led to a large diversity of CI instances. While specialized, competing and alternative CI building blocks are inherent to a healthy ecosystem, it also becomes apparent that the increasing degree of fragmentation is hindering interoperation, and thus limiting collaboration, which is essential for modern science communities often spanning international groups and multiple disciplines (but even 'small sciences', with smaller and localized communities, are often embedded into the larger scientific ecosystem, and are increasingly dependent on the availability of CI.) There are different reasons why fragmentation occurs, on technical and social level. But also, it is apparent that the current funding model for creating CI components largely fails to aid the transition from research to production, by mixing CS research and IT engineering challenges into the same funding strategies. The 10 th anniversary of the EU funded project 'Grid Lab' (which was an early and ambitious attempt on providing a consolidated and science oriented cyberinfrastructure software stack to a specific science community) was taken as an opportunity to invite international leaders and early stage researchers in grid computing and e-Science from Europe, America and Asia, and, together with representatives of the EU and US funding agencies, to discuss the fundamental aspects of CI evolution, and to contemplate the options for a more coherent, more coordinated approach to the global evolution of CI.
Building an infrastructure for scientific Grid computing: status and goals of the EGEE project
Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2005
The state of computer and networking technology today makes the seamless sharing of computing resources on an international or even global scale conceivable. Scientific computing Grids that integrate large, geographically distributed computer clusters and data storage facilities are being developed in several major projects around the world. This article reviews the status of one of these projects, Enabling Grids for E-SciencE, describing the scientific opportunities that such a Grid can provide, while illustrating the scale and complexity of the challenge involved in establishing a scientific infrastructure of this kind.
Interoperable execution of eScience applications on Grids & Clouds through open standards
2012
The advent of cloud computing has offered scientific communities the ability to access computational resources that can satisfy their growing research needs starting to outgrow the size of traditional local resources as PCs and locally managed clusters and also of grid sites. Since grids and clouds are heterogeneous in nature and are based on different middlewares, interoperability between the service interfaces exposing the capabilities of these infrastructures is recognized as an important issue. This problem is usually handled by using the appropriate adaptors to interact with several middlewares thus allowing the applications to be executed on federated infrastructures. While aiming for federated resources access, there is an overhead for the application clients to continuously detect and adapt to every evolution of the target middlewares. In the presented work a complementary approach is followed to circumvent this problem by enabling interoperability between different execution services through the adoption of open and widely adopted standards.
2005
The Application Hosting Environment (AHE) is designed to help the scientist who wants to use Grid [1-3] computing by providing a higher level abstraction of the Grid than is currently offered by existing Grid middleware such as Globus [4]. Rather than providing capabilities for submitting generic computational jobs to a Grid we provide the scientist with services that will start and manage the applications he wants to use on his behalf; we do this by building a layer of middleware on top of the existing grid middleware. By providing this extra level of abstraction the scientist does not need to know the details of any particular grid middleware, so we can isolate him from any changes to the underlying grid middleware used by a particular grid.
Grid and P2P Middleware for Scientific Computing Systems
2010 International Conference on Complex, Intelligent and Software Intensive Systems, 2010
Grid and P2P systems have achieved a notable success in the domain of scientific and engineering applications, which commonly demand considerable amounts of computational resources. However, Grid and P2P systems remain still difficult to be used by the domain scientists and engineers due to the inherent complexity of the corresponding middleware and the lack of adequate documentation. In this paper we survey recent developments of Grid and P2P middleware in the context of scientific computing systems. The differences on the approaches taken for Grid and P2P middleware as well as the common points of both paradigms are highlighted. In addition, we discuss the corresponding programming models, languages, and applications.