arXiv subjects

Ruben M. Cabezon

Publications and source records attributed to Ruben M. Cabezon.

6 recordsLinked to original sources

arXiv2019

Two-level Dynamic Load Balancing for High Performance Scientific Applications

Scientific applications are often complex, irregular, and computationally-intensive. To accommodate the ever-increasing computational demands of scientific applications, high-performance computing (HPC) systems have become larger and more complex, offering parallelism at multiple levels (e.g., nodes, cores per node, threads per core). Scientific applications need to exploit all the available multilevel hardware parallelism to harness the available computational power. The performance of applications executing on such HPC systems may adversely be affected by load imbalance at multiple levels, caused by problem, algorithmic, and systemic characteristics. Nevertheless, most existing load balancing methods do not simultaneously address load imbalance at multiple levels. This work investigates the impact of load imbalance on the performance of three scientific applications at the thread and process levels. We jointly apply and evaluate selected dynamic loop self-scheduling (DLS) techniques to both levels. Specifically, we employ the extended LaPeSD OpenMP runtime library at the thread level and extend the DLS4LB MPI-based dynamic load balancing library at the process level. This approach is generic and applicable to any multiprocess-multithreaded computationally-intensive application (programmed using MPI and OpenMP). We conduct an exhaustive set of experiments to assess and compare six DLS techniques at the thread level and eleven at the process level. The results show that improved application performance, by up to 21%, can only be achieved by jointly addressing load imbalance at the two levels. We offer insights into the performance of the selected DLS techniques and discuss the interplay of load balancing at the thread level and process level.

cs.DC↗

arXiv2017

SPHYNX: an accurate density-based SPH method for astrophysical applications

Hydrodynamical instabilities and shocks are ubiquitous in astrophysical scenarios. Therefore, an accurate numerical simulation of these phenomena is mandatory to correctly model and understand many astrophysical events, such as Supernovas, stellar collisions, or planetary formation. In this work, we attempt to address many of the problems that the smoothed particle hydrodynamics (SPH) technique has when dealing with subsonic hydrodynamical instabilities or shocks. To that aim we built a new SPH code named SPHYNX, that includes many of the recent advances in the SPH technique and some other new ones, which we present here. SPHYNX is of Newtonian type and grounded in the Euler-Lagrange formulation of the SPH technique. Its distinctive features are: the use of an integral approach to estimating the gradients; the use of a flexible family of interpolators called sinc kernels, which suppress pairing instability; and the incorporation of a new type of volume element which provides a better partition of the unity. Unlike other modern formulations, which consider volume elements linked to pressure, our volume element choice relies on density. SPHYNX conserves mass, linear and angular momentum, energy, entropy, and preserves kernel normalization even in strong shocks. The coupling between the integral approach to calculate gradients and the new family of volume elements reduces the so-called tensile instability. Both features help to suppress the damp which often prevents the growth of hydrodynamic instabilities in regular SPH codes. On the whole, SPHYNX has passed the verification tests described below. For identical particle setting and initial conditions the results were similar (or better in some particular cases) than those obtained with other SPH schemes such as GADGET-2, PSPH or with the recent density-independent formulation (DISPH) and conservative reproducing kernel (CRKSPH) techniques.

astro-ph.IM↗

arXiv2012

Testing the concept of integral approach to derivatives within the smoothed particle hydrodynamics technique in astrophysical scenarios

The behavior of IAD_0 scheme, a fully conservative SPH scheme based on a tensor formulation, is analyzed in connection with several astrophysical scenarios, and compared to the same simulations carried out with the standard SPH technique. The proposed hydrodynamic scheme is validated using a variety of numerical tests that cover important topics in astrophysics, such as the evolution of supernova remnants, the stability of self-gravitating bodies and the coalescence of compact objects. The results suggest that the SPH scheme built with the integral approach to the derivatives premise improves the results of the standard SPH technique. In particular, it is observed a better development of hydrodynamic instabilities, an improved description of self-gravitant structures in equilibrium and a reasonable description of the process of coalescence of two white dwarfs. A good energy, and linear and angular momentum conservation, generally better than that of standard SPH, was also obtained. In addition the new scheme is less susceptible to suffer pairing instability.

astro-ph.IM↗

arXiv2011

Improving smoothed particle hydrodynamics with an integral approach to calculating gradients

In this paper we develop and test a fully conservative SPH scheme based on a tensor formulation that can be applied to simulate astrophysical systems. In the proposed scheme, derivatives are calculated from an integral expression that leads to a tensor (instead of a vectorial) estimation of gradients and reduces to the standard formulation in the continuum limit. The new formulation improves the interpolation of physical magnitudes, leading to a set of conservative equations that resembles those of standard SPH. The resulting scheme is verified using a variety of well-known tests, all of them simulated in two dimensions. We also discuss an application of the proposed tensor method to astrophysics by simulating the stability of a Sun-like polytrope calculated in three dimensions.

astro-ph.IM↗

arXiv2009

Pulsating reverse detonation models of Type Ia supernovae. II: Explosion

Observational evidences point to a common explosion mechanism of Type Ia supernovae based on a delayed detonation of a white dwarf. However, all attempts to find a convincing ignition mechanism based on a delayed detonation in a destabilized, expanding, white dwarf have been elusive so far. One of the possibilities that has been invoked is that an inefficient deflagration leads to pulsation of a Chandrasekhar-mass white dwarf, followed by formation of an accretion shock that confines a carbon-oxygen rich core, while transforming the kinetic energy of the collapsing halo into thermal energy of the core, until an inward moving detonation is formed. This chain of events has been termed Pulsating Reverse Detonation (PRD). In this work we present three dimensional numerical simulations of PRD models from the time of detonation initiation up to homologous expansion. Different models characterized by the amount of mass burned during the deflagration phase, M_defl, give explosions spanning a range of kinetic energies, K ~ (1.0-1.2) foes, and 56Ni masses, M(56Ni) ~ 0.6-0.8 M_sun, which are compatible with what is expected for typical Type Ia supernovae. Spectra and light curves of angle-averaged spherically symmetric versions of the PRD models are discussed. Type Ia supernova spectra pose the most stringent requirements on PRD models.

astro-ph.SR↗

arXiv2008

A one-parameter family of interpolating kernels for Smoothed Particle Hydrodynamics studies

A set of interpolating functions of the type f(v)={(sin[v pi/2])/(v pi/2)}^n is analyzed in the context of the smoothed-particle hydrodynamics (SPH) technique. The behaviour of these kernels for several values of the parameter n has been studied either analytically as well as numerically in connection with several tests carried out in two dimensions. The main advantage of this kernel relies in its flexibility because for n=3 it is similar to the standard widely used cubic-spline, whereas for n>3 the interpolating function becomes more centrally condensed, being well suited to track discontinuities such as shock fronts and thermal waves.

astro-ph↗