Robert C. Armstrong

A Component Architecture for High-Performance Scientific Computing

The Common Component Architecture (CCA) provides a means for software developers to manage the complexity of large-scale scientific simulations and to move toward a plug-and-play environment for high-performance coputing. In the scientific computing context, component models also promote collaboration using independently developed software, thereby allowing particular individals or groups to focus on the aspects of greatest interest to them. The CCA supports parallel and distributed coputing as well as local high-performance connections between components in a language-independent manner. The design places minimal requirements on components and thus facilitates the integration of existing code into the CCA environment. The CCA model imposes minimal ovehead to minimize the impact on application performance. The focus on high performance distinguishes the CCA from most other component models. The CCA is being applied within an increasing range of disciplines, including cobustion research, global climate simulation, and computtional chemistry.

The CCA core specification in a distributed memory SPMD framework

We present an overview of the Common Component Architecture (CCA) core specification and CCAFFEINE, a Sandia National Laboratories framework implementation compliant with the draft specification. CCAFFEINE stands for CCA Fast Framework Example In Need of Everything; that is, CCAFFEINE is fast, lightweight, and it aims to provide every framework service by using external, portable components instead of integrating all services into a single, heavy framework core. By fast, we mean that the CCAFFEINE glue does not get between components in a way that slows down their interactions. We present the CCAFFEINE solutions to several fundamental problems in the application of component software approaches to the construction of single program multiple data (SPMD) applications. We demonstrate the integration of components from three organizations, two within Sandia and one at Oak Ridge National Laboratory. We outline some requirements for key enabling facilities needed for a successful component approach to SPMD application building. Copyright

Models for gasless combustion in layered materials and random media

Abstract It is generally assumed in this work, as elsewhere, that the mass diffusion coefficient has an Arrhenius temperature dependence. This allows a combustion wave propagated by thermal conduction to develop similar to more conventional combustion systems. It is found that for realistic systems the problem must involve a three length scale analysis. Similar to the classic theory on premixed, gas-phase flames, the largest scale is identified with thermal conduction and a “reaction zone” scale proportional to the inverse of the mass diffusion activation energy is associated with a small length over which significant mass diffusion is possible. Unlike classic flame theory a still smaller scale is identified with the length over which mass diffusion takes place (i.e., the size of a typical domain of alloyable constituent). Flame speeds are derived for three different geometric configurations of a binary system using a singular perturbation analysis. First a system where the constituents are arrayed in alt...

Influences of two-phase flow in the deflagration of homogeneous solids

Theoretical analyses are developed for the deflagration of solids such as nitramines that experience exothermic reactions in liquid layers at their surfaces. Relative motion of gas and liquid in a two-phase region at the surface is considered, with influences of pressure gradients and of surface-tension gradients taken into account for the drops and bubbles. It is shown that these influences tend to produce gas velocities in excess of liquid velocities. Burning-rate expressions are derived by activation-energy asymptotics, with special attention paid to the role of interphase heat transfer.

Computational quality of service for scientific components.

Scientific computing on massively parallel computers presents unique challenges to component-based software engineering (CBSE). While CBSE is at least as enabling for scientific computing as it is for other arenas, the requirements are different. We briefly discuss how these requirements shape the Common Component Architecture, and we describe some recent research on quality-of-service issues to address the computational performance and accuracy of scientific simulations.

Two Asymptotic Models for Solid Propellant Combustion

Abstract —We derive two different asymptotic models which describe the nonsteady, nonplanar burning of certain types of homogeneous solid propellants. Motivated in part by recent work on ammonium perchlorate deflagration, we assume, in the first model, that a fraction of the pro-pellant is pyrolyzed directly to product gases at a solid/gas interface, while the remainder sublimes and burns in the gas phase. In the second model, there is a thin liquid layer between the solid and gas, with combustion occurring in both the liquid and gas phases. Our analysis exploits the largeness of activation energies to derive flame sheet models analogous to those derived for strictly gaseous and strictly condensed deflagrations. For the special case of steady, planar burning, we obtain expressions for the regression rate eigenvalue as a function of the various parameters in the problem. However, a linear stability analysis of this basic solution shows that, for sufficiently large values of a certain grouping of parameters...

Parallel PDE-Based Simulations Using the Common Component Architecture

The complexity of parallel PDE-based simulations continues to increase as multimodel, multiphysics, and multi-institutional projects become widespread. A goal of component- based software engineering in such large-scale simulations is to help manage this complexity by enabling better interoperability among various codes that have been independently developed by different groups. The Common Component Architecture (CCA) Forum is defining a component architecture specification to address the challenges of high-performance scientific computing. In addition, several execution frameworks, supporting infrastructure, and general-purpose components are being developed. Furthermore, this group is collaborating with others in the high-performance computing community to design suites of domain-specific component interface specifications and underlying implementations.

Theoretical models for the combustion of alloyable materials

The purpose of this work is to extend a theoretical model of layered (laminar) media for SHS combustion presented in an earlier article [1] to explore possible mechanisms for after-burning in SHS (i.e., gasless) combustion. As before, our particular interest is how the microscopic geometry of the solid reactants is reflected in the combustion wave and in the reaction product. The model is constructed from alternating lamina of two pure reactants that interdiffuse exothermically to form a product. Here, the laminar model is extended to contain layers of differing thicknesses. Using asymptotic theory, it was found that under certain conditions, the combustion wave can become “detached,” and an initial thin flame propagates through the media, leaving a slower, thicker flame following behind (i.e., afterburning). Thin laminae are consumed in the initial flame and are thick in the secondary. The thin flame has a width determined by the inverse of the activation energy of diffusion, as found previously. The width of the afterburning zone, however, is determined by the absolute time of diffusion for the thicker laminae. Naturally, when the laminae are all the same thickness, there is only one thin flame. The condition for the appearance of afterburning is found to be contingent on the square of the ratio of smallestto-largest thicknesses being considerably less than unity.

Performance measurement and modeling of component applications in a high performance computing environment: a case study

Summary form only given. We present a case study of performance measurement and modeling of a CCA (common component architecture) component-based application in a high performance computing environment. Component-based HPC applications allow the possibility of creating component-level performance models and synthesizing them into application performance models. However, they impose the restriction that performance measurement/monitoring needs to be done in a nonintrusive manner and at a fairly coarse-grained level. We propose a performance measurement infrastructure for HPC based loosely on recent work done for grid environments. A prototypical implementation of the infrastructure is used to collect data for three components in a scientific application and construct their performance models. Both computational and message-passing performance are addressed.

Performance technology for parallel and distributed component software

This work targets the emerging use of software component technology for high‐performance scientific parallel and distributed computing. While component software engineering will benefit the construction of complex science applications, its use presents several challenges to performance measurement, analysis, and optimization. The performance of a component application depends on the interaction (possibly nonlinear) of the composed component set. Furthermore, a component is a ‘binary unit of composition’ and the only information users have is the interface the component provides to the outside world. A performance engineering methodology and development approach is presented to address evaluation and optimization issues in high‐performance component environments. We describe a prototype implementation of a performance measurement infrastructure for the Common Component Architecture (CCA) system. A case study demonstrating the use of this technology for integrated measurement, monitoring, and optimization in CCA component‐based applications is given. Copyright