Publications Search

A new adaptive tabulation scheme for multi-phase equations of state (EOS) is described. Adaptation allows verification that a table represents an EOS model to some desired accuracy at a much lower computational cost than standard tables. Computational efficiency is provided through the use of a quad-tree representation. Using both rectangular and triangular interpolation regions results in accurate descriptions of phase boundaries. The new format is demonstrated on a representative multi-phase EOS model. © 2012 American Institute of Physics.

More Details

TYPE Conference YEAR 2012

OSTI Scopus

Molecular dynamics simulation of dynamic response of beryllium

AIP Conference Proceedings

Thompson, Aidan P.; Lane, James M.; Desjarlais, Michael P.

The response of beryllium to dynamic loading has been extensively studied, both experimentally and theoretically, due to its importance in several technological areas. We use a MEAM empirical potential to examine the melt transition. MD simulations of equilibrated two-phase systems were used to calculate the HCP melting curve up to 300 GPa. This was found to agree well with previous ab initio calculations. The Hugoniostat method was used to examine dynamic compression along the two principal orientations of the HCP crystal. In both directions, the melting transition occurred at 230 GPa and 5000 K, consistent with the equilibrium melting curve. Direct NEMD simulations of uniaxial compression show a transition to an amorphous material at shocked states that lie below the equilibrium melt curve. © 2012 American Institute of Physics.

More Details

TYPE Journal Article YEAR 2012

OSTI Scopus

Optimization-based modeling with applications to transport: Part 3. Computational studies

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Ridzal, Denis R.; Young, Joseph G.; Bochev, Pavel B.; Peterson, Kara J.

This paper is the final of three related articles that develop and demonstrate a new optimization-based framework for computational modeling. The framework uses optimization and control ideas to assemble and decompose multiphysics operators and to preserve their fundamental physical properties in the discretization process. One application of the framework is in the formulation of robust algorithms for optimization-based transport (OBT). Based on the theoretical foundations established in Part 1 and the optimization algorithm for the solution of the remap subproblem, derived in Part 2, this paper focuses on the application of OBT to a set of benchmark transport problems. Numerical comparisons with two other transport schemes based on incremental remapping, featuring flux-corrected remap and the linear reconstruction with van Leer limiting, respectively, demonstrate that OBT is a competitive transport algorithm. © 2012 Springer-Verlag.

More Details

TYPE Conference YEAR 2012

OSTI Scopus

Towards Automated Memory Model Generation Via Event Tracing

Computer Journal

Hammond, Simon D.

The importance of memory performance and capacity is a growing concern for high performance computing laboratories around the world. It has long been recognized that improvements in processor speed exceed the rate of improvement in dynamic random access memory speed and, as a result, memory access times can be the limiting factor in high performance scientific codes. The use of multi-core processors exacerbates this problem with the rapid growth in the number of cores not being matched by similar improvements in memory capacity, increasing the likelihood of memory contention. In this paper, we present WMTools , a lightweight memory tracing tool and analysis framework for parallel codes, which is able to identify peak memory usage and also analyse per-function memory use over time. An evaluation of WMTools , in terms of its effectiveness and also its overheads, is performed using nine established scientific applications/benchmark codes representing a variety of programming languages and scientific domains. We also show how WMTools can be used to automatically generate a parameterized memory model for one of these applications, a two-dimensional non-linear magnetohydrodynamics application, Lare2D . Through the memory model we are able to identify an unexpected growth term which becomes dominant at scale. With a refined model we are able to predict memory consumption with under 7% error.

More Details

TYPE Journal Article YEAR 2012

OSTI DOI