As a result of this award NSF will partially fund an international workshop on Clusters, Clouds and Grids for Scientific Computing. This workshop will explore the transformation taking place through the combined used of clusters, clouds and grounds in the context of multicore and hybrid microprocessor designs, escalating volumes of data and by the spread of computational and data intensive methods across nearly every research domain. The topic is relevant to the cyberinfrastructure for U.S. science and engineering researchers.
The research areas of cluster, cloud and grid computing, which today provide fundamental infrastructure for all areas of advanced computational science, are being radically transformed by the convergence of at least two unprecedented trends. First, the components that make up high performance computers, and computers in general, are changing drastically. Where a decade ago, the majority of a computer’s performance was derived from a CPU, now most computers—and especially supercomputers—utilize CPUs with multiple processing cores and separate co-processors like GPUs, creating a "hybrid" system architecture. As one can imagine, the programming of applications has to change just as drastically to function on these new hybrid systems, leaving a significant challenge to port legacy code to new and upcoming technology platforms. Second, the amount of data being generated by today’s scientific research is unprecedented. This big data is challenging not only because of the sheer volume of bits that must be processed and managed, but also because of the logistical problems associated with making the data of most current interest available to participants in large national and international collaborations, sitting in different administrative domains, spread across the wide area network, and wanting to use diverse resources like clusters, clouds and grids. For the cyberinfrastructure research community, these new conditions raise a broad set of challenging design and deployment issues in areas such as scalability, programmability, performance, interoperability, resilience, resource virtualization, data logistics and system management. The Workshop on Clusters, Clouds and Grids for Scientific Computing (CCGSC) described here helps address this unprecedented situation by providing a forum for creative dialogue between leading edge researchers in the field in order to produce the kind of coordinated review and penetrating analysis that can help translate such momentous developments into momentous benefits for the scientific community. The workshop format was organized to maximize the synergistic effects of combining researchers in all the focus areas together in one workshop, so as to initiate new directions for high impact research that require coordination in two or more of these areas. To achieve this goal, the workshop concentrated on the following tasks: Survey and analyze the key deployment, operational and usage issues for clusters, clouds and grids, especially focusing on discontinuities produced by multicore and hybrid architectures, data intensive science, and the increasing need for wide area/local area interaction. Document the current state-of-the-art in each of these areas, identifying interesting questions and limitations. Experiences with clusters, clouds and grids relative to their science research communities and science domains that are benefitting from the technology. Explore interoperability among disparate clouds as well as interoperability between various clouds and grids and the impact on the domain sciences. Explore directions for future research and development against the background of disruptive trends and technologies and the recognized gaps in the current state-of-the-art. This workshop was one in a series of biennial meetings, going back over twenty years. The series represents collaboration between U.S. and French colleagues who are well situated to recruit outstanding international participants from the field and to leverage the resulting ideas, insights, and innovations to benefit the broader research community. While providing a model of what such international collaborations should be, we believe this series has helped lay the groundwork for the international partnerships in hardware and software cyberinfrastructure that are becoming more and more critical to rapid progress in the fields of science and engineering, and are now vital to the long term welfare of our society.
