Search NASA⌕ Search

Engineering topics

Abbasi, H

Publications and source records attributed to Abbasi, H.

In Situ Methods, Infrastructures, and Applications on High Performance Computing Platforms, a State-of-the-art (STAR) Report

The considerable interest in the high performance computing (HPC) community regarding analyzing and visualization data without first writing to disk, i.e., in situ processing, is due to several factors. First is an I/O cost savings, where data is analyzed /visualized while being generated, without first storing to a filesystem. Second is the potential for increased accuracy, where fine temporal sampling of transient analysis might expose some complex behavior missed in coarse temporal sampling. Third is the ability to use all available resources, CPU’s and accelerators, in the computation of analysis products. This STAR paper brings together researchers, developers and practitioners using in situ methods in extreme-scale HPC with the goal to present existing methods, infrastructures, and a range of computational science and engineering applications using in situ analysis and visualization.

97 MATHEMATICS AND COMPUTING↗

Hybrid (particle in cell-fluid) simulation of ion-acoustic soliton generation including super-thermal and trapped electrons

The present paper is devoted to the simulation of the nonlinear disintegration of a localized perturbation into an ion-acoustic soliton in a plasma. Recently, this problem was studied by a simple model [H. Abbasi et al., Plasma Phys. Controlled Fusion 50, 095007 (2008)]. The main assumptions were (i) in the electron velocity distribution function (DF), the ion-acoustic soliton velocity was neglected in comparison to the electron thermal velocity, (ii) on the ion-acoustic evolution time-scale, the electron velocity DF was assumed to be stationary, and (iii) the calculation was restricted to the small amplitude case. In order to generalize the model, one has to consider the evolution of the electron velocity DF for finite amplitudes. For this purpose, a one dimensional electrostatic hybrid code, particle in cell (PIC)-fluid, was designed. It simulates the electrons dynamics by the PIC method and the cold ions dynamics by the fluid equations. The plasma contains a population of super-thermal electrons and, therefore, a Lorentzian (kappa) velocity DF is used to model the high energy tail in the electron velocity DF. Electron trapping is included in the simulation in view of their nonlinear resonant interaction with the localized perturbation. A Gaussian initial perturbation is used to model the localized perturbation. The influence of both the trapped and the super-thermal electrons on this process is studied and compared with the previous model.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Fully kinetic simulation of ion acoustic and dust-ion acoustic waves

A series of numerical simulations is presented, based on a recurrence-free Vlasov kinetic model using kinetic phase point trajectories. All plasma components are modeled kinetically via a Vlasov evolution equation, then coupled through Poisson's equation. The dynamics of ion acoustic waves in an electron-ion and in a dusty (electron-ion-dust) plasma configuration are investigated, focusing on wave decay due to Landau damping and, in particular, on the parametric dependence of the damping rate on the dust concentration and on the electron-to-ion temperature ratio. In the absence of dust, the occurrence of damping was observed, as expected, and its dependence to the relative magnitude of the electron vs ion temperature(s) was investigated. When present, the dust component influences the charge balance, enabling dust-ion acoustic waves to survive Landau damping even in the extreme regime where T{sub e} {approx_equal} T{sub i}. The Landau damping rate is shown to be minimized for a strong dust concentration or/and for a high value of the electron-to-ion temperature ratio. Our results confirm earlier theoretical considerations and contribute to the interpretation of experimental observations of dust-ion acoustic wave characteristics.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Modulational instability of ion-acoustic waves in plasmas with superthermal electrons

Using the reductive perturbation technique, the modulational instability of ion- acoustic waves in a plasma containing superthermal electrons is studied. It is found that the presence of superthermal electrons significantly changes the instability domain. A Lorentzian (kappa) velocity distribution function is used to model superthermal electrons. It is shown that the presence of superthermal electrons reduces the critical frequency of the modulational instability of ion-acoustic waves. Besides, due to the presence of the superthermal electrons, ion-acoustic waves are unstable on a vaster region. Moreover, the modulational instability growth rate is larger for a larger population of superthermal electrons.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Skel: Generative Software for Producing Skeletal I/O Applications

Massively parallel computations consist of a mixture of computation, communication, and I/O. As part of the co-design for the inevitable progress towards exascale computing, we must apply lessons learned from past work to succeed in this new age of computing. Of the three components listed above, implementing an effective parallel I/O solution has often been overlooked by application scientists and was usually added to large scale simulations only when existing serial techniques had failed. As scientists teams scaled their codes to run on hundreds of processors, it was common to call on an I/O expert to implement a set of more scalable I/O routines. These routines were easily separated from the calculations and communication, and in many cases, an I/O kernel was derived from the application which could be used for testing I/O performance independent of the application. These I/O kernels developed a life of their own used as a broad measure for comparing different I/O techniques. Unfortunately, as years passed and computation and communication changes required changes to the I/O, the separate I/O kernel used for benchmarking remained static no longer providing an accurate indicator of the I/O performance of the simulation making I/O research less relevant for the application scientists. In this paper we describe a new approach to this problem where I/O kernels are replaced with skeletal I/O applications automatically generated from an abstract set of simulation I/O parameters. We realize this abstraction by leveraging the ADIOS middleware's XML I/O specification with additional runtime parameters. Skeletal applications offer all of the benefits of I/O kernels including allowing I/O optimizations to focus on useful I/O patterns. Moreover, since they are automatically generated, it is easy to produce an updated I/O skeleton whenever the simulation's I/O changes. In this paper we analyze the performance of automatically generated I/O skeletal applications for the S3D and GTS codes. We show that these skeletal applications achieve performance comparable to that of the production applications. We wrap up the paper with a discussion of future changes to make the skeletal application better approximate the actual I/O performed in the simulation.

99 GENERAL AND MISCELLANEOUS↗

Influence of superthermal and trapped electrons on oblique propagation of ion-acoustic waves in magnetized plasma

Effects of superthermal and trapped electrons on the oblique propagation of linear and nonlinear ion-acoustic waves in an electron-ion plasma in the presence of a uniform external magnetic field are investigated. In order to model the superthermal electrons, a Lorentzian (kappa) velocity distribution function has been employed. The ions are cold and their dynamics are studied by hydrodynamic equations. First, the linear dispersion relation of the fast and slow modes are obtained. It is shown that the superthermal electrons cause the both modes to propagate with smaller phase velocities. Then, modified Korteweg-de Vries equations describing the propagation of nonlinear slow and fast ion-acoustic waves are derived. It is shown that the presence of superthermal and trapped electrons has great influence on the nature of magnetized ion-acoustic solitons. The dependency of soliton attributes to the parameters associated with the superthermality and trapping mechanism will be shown.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Influence of electron velocity distribution on the plasma expansion features

Collisionless plasma expansion into vacuum is addressed emphasizing on the kinetic effects associated with the plasma electrons. It is an important issue since there are situations in which the plasmas are in nonequilibrium state. Thus, the electron distribution function (DF) that is generally non-Maxwellian has to be modeled. For this purpose, the generalized Lorentzian (kappa) DF is used to simulate the electron DF. The Maxwellian and kappa distributions differ substantially in a high-energy tail. Thus, the electron dynamics is studied by the Vlasov equation. Neglecting the ion temperatures, fluid equations are used for them. It is shown that by increasing the population of energetic electrons, the expansion takes place faster, the resulting electric field is stronger, and the ions are accelerated to higher energy.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Extending I/O through High Performance Data Services

The complexity of HPC systems has increased the burden on the developer as applications scale to hundreds of thousands of processing cores. Moreover, additional efforts are required to achieve acceptable I/O performance, where it is important how I/O is performed, which resources are used, and where I/O functionality is deployed. Specifically, by scheduling I/O data movement and by effectively placing operators affecting data volumes or information about the data, tremendous gains can be achieved both in the performance of simulation output and in the usability of output data. Previous studies have shown the value of using asynchronous I/O, of employing a staging area, and of performing select operations on data before it is written to disk. Leveraging such insights, this paper develops and experiments with higher level I/O abstractions, termed 'data services', that manage output data from 'source to sink': where/when it is captured, transported towards storage, and filtered or manipulated by service functions to improve its information content. Useful services include data reduction, data indexing, and those that manage how I/O is performed, i.e., the control aspects of data movement. Our data services implementation distinguishes control aspects - the control plane - from data movement - the data plane, so that both may be changed separably. This results in runtime flexibility not only in which services to employ, but also in where to deploy them and how they use I/O resources. The outcome is consistently high levels of I/O performance at large scale, without requiring application change.

97 MATHEMATICS AND COMPUTING↗