Jose Garcia

Ontology-supported scientific data frameworks: The Virtual Solar-Terrestrial Observatory experience

We have developed a semantic data framework that supports interdisciplinary virtual observatory projects across the fields of solar physics, space physics and solar-terrestrial physics. This work required a formal, machine understandable representation for concepts, relations and attributes of physical quantities in the domains of interest as well as their underlying data representations. To fulfill this need, we developed a set of solar-terrestrial ontologies as formal encodings of the knowledge in the Ontology Web Language-Description Logic (OWL-DL) format. We present our knowledge representation and reasoning needs motivated by the context of Virtual Observatories, from fields spanning upper atmospheric terrestrial physics to solar physics, whose intent is to provide access to observational datasets. The resulting data framework is built upon semantic web methodologies and technologies and provides virtual access to distributed and heterogeneous sets of data as if all resources appear to be organized, stored and retrieved from a local environment. Our conclusion is that the combination of use case-driven, small and modular ontology development, coupled with free and open-source software tools and languages provides sufficient expressiveness and capabilities for an initial production implementation and sets the stage for a more complete semantic-enablement of future frameworks.

Semantically-enabled large-scale science data repositories

Large heterogeneous online repositories of scientific information have the potential to change the way science is done today. In order for this potential to be realized, numerous challenges must be addressed concerning access to and interoperability of the online scientific data. In our work, we are using semantic web technologies to improve access and interoperability by providing a framework for collaboration and a basis for building and distributing advanced data simulation tools. Our initial scientific focus area is the solar terrestrial physics community. In this paper, we will present our work on the Virtual Solar Terrestrial Observatory (VSTO). We will present the emerging trend of the virtual observatory – a virtual integrated evolving scientific data repository – and describe the general use case and our semantically-enabled architecture. We will also present our specific implementation and describe the benefits of the semantic web in this setting. Further, we speculate on the future of the growing adoption of semantic technologies in this important application area of scientific cyberinfrastructure and semantically enabled scientific data repositories.

Enabling Scientific Research using an Interdisciplinary Virtual Observatory: The Virtual Solar-Terrestrial Observatory Example

Our work is aimed at enabling a new style of virtual, distributed scientific research. We have designed, built, and deployed an interdisciplinary virtual observatory—an online service providing access to what appears to be an integrated collection of scientific data. The Virtual Solar-Terrestrial Observatory (VSTO) is a production semantic web data framework providing access to observational data sets from fields spanning upper atmospheric terrestrial physics to solar physics. The observatory allows virtual access to a highly distributed and heterogeneous set of data that appears as if all resources are organized, stored, and retrieved or used in a common way. The end-user community includes scientists, students, and data providers. We will introduce interdisciplinary virtual observatories and their potential impact by describing our experiences with VSTO. We will also highlight some benefits of the embedded semantic web technology and also provide evaluation results after the first year of use.

GPU Acceleration of a Cloud Resolving Model using CUDA

Abstract The increasing computing power of graphics processing units (GPU) has motivated the use of GPUs to speed up climate models. Their low price in conjunction with C style parallel programming tools, like CUDA (Compute Unified Device Architecture), allow the programmers, even non-programmers, to exploit the fine grain parallelism power of GPU conveniently. In this paper, we report successful acceleration of a Cloud Resolving Model (CRM) by implementing the MPDATA algorithm on the GPU using CUDA. The results show the great speedup potential for GPU acceleration of CRM which could carry over to other atmospheric models.

Parallel adaptive mesh refinement for first-order system least squares

SUMMARY n nThis paper develops new adaptive mesh refinement strategies for first-order system least squares (FOSLS) in conjunction with algebraic multigrid (AMG) methods in the context of nested iteration (NI). The goal is to reach a certain error tolerance with the least amount of computational cost and nearly uniform distribution of the error over all elements. To accomplish this, the refinement decisions at each refinement level are determined on the basis of minimizing the ‘accuracy-per-computational-cost’ efficiency (ACE) measure that takes into account both error reduction and computational cost. The NI-FOSLS-AMG-ACE approach produces a sequence of refinement levels in which the error is equally distributed across elements on a relatively coarse grid. Once the solution is numerically resolved, refinement becomes nearly uniform. Accommodations of the ACE approach to massively distributed memory architectures involve a geometric binning strategy to reduce communication cost. Load balancing begins at very coarse levels. Elements and nodes are redistributed using parallel quadtree structures and a space-filling curve, which automatically ameliorates load balancing issues at finer levels. Numerical results show that the NI-FOSLS-AMG-pACE approach is able to provide highly accurate approximations to rapidly varying solutions at relatively low cost. Excellent weak and strong scalability are demonstrated on 4096 processors for problems with 15 million biquadratic elements.Copyright