ExMatEx Overview DoE Exascale Co-Design Center for Materials in Extreme Environments.
The objective of the Exascale Co-design Center for Materials in Extreme Environments (ExMatEx) is to establish the interrelationship among algorithms, system software, and hardware required to develop a multiphysics exascale simulation framework for modeling materials subjected to extreme mechanical and radiation environments. Such a simulation capability will play a key role in solving many of today’s most pressing problems, including producing clean energy, extending nuclear reactor lifetimes, and certifying the aging nuclear stockpile.
Our goal is to establish the interrelationships between hardware, middleware (software stack), programming models and algorithms to enable a productive exascale environment for multiphysics simulations of materials in extreme mechanical and radiation environments.
We will exploit, rather than avoid, the greatly increased levels of concurrency, heterogeneity, and flop/byte ratios expected on the upcoming extreme scale platforms.
Co-design of the exascale ecosystem involves ExMatEx and other application co-design centers working in concert with hardware vendors and other computer science research activities as illustrated in the image to the right. The computer system architecture is shown to contain both the emerging hardware as well as all of the software stack required to operate the computer. The role of the application co-design center is to both introduce our evolving application requirements and workflow into the exascale ecosystem through proxy application and to evaluate these applications in the context of emerging hardware and software solutions.
Research Areas Our research spans a broad set of topics including computer science, algorithms, and applications.
We're engaged in the assessment and evaluation of both traditional programming languages, emering programming paradigms, and domain specific languages.
Runtime system services can support “programming in the large”—coupling multiple diverse components of a dynamic multi-scale computation and orchestrating its execution on the system.
Hardware-interfacing tools combining analytical models, architectural simulation, system emulation and empirical measurements drive the co-design process.
Proxy Applications We release most of our software as open source.
Our proxy applications, and the algorithms they implement, are the primary vehicle for collaboration with both vendor and academic partners. They are also heavily used within our center to explore programming models, systemware, and hardware-interfacing tools.
CoMD is a reference implementation of classical molecular dynamics algorithms and workloads as used in materials science. The code is intended to serve as a vehicle for co-design by allowing others to extend and/or reimplement it as needed to test performance of new architectures, programming models, etc. New versions of CoMD will be released to incorporate the lessons learned from the co-design process.
CoGL is a meso-scale simulation proxy app used to analyze pattern formation in ferroelastic materials using the Ginzburg–Landau approach. It models transitions from a face-centered cubic parent phase to a body-centered tetragonal product phase due to an external deformation. The implementation uses a data-parallel approach, making use of the PISTON framework, developed at Los Alamos as an extension of NVIDIA’s Thrust library.
VPFFT is an implementation of a mesoscale micromechanical materials model. By solving the viscoplasticity model, VPFFT simulates the evolution of a material under deformation. The solution time to the viscoplasticity model, described by a set of partial differential equations, is significantly reduced by the application of Fast Fourier Transform in the VPFFT algorithm.
CoHMM represents the basic workflow requirements of a scale-bridging application. The coarse- and fine-scale models, and the “glue” connecting the two, are intentionally kept simple to focus attention on the dataflow and workload requirements of a task-based scale-bridging model, enabling experimentation with a variety of programming models and runtime systems.
LULESH, developed at LLNL as part of the DARPA UHPC prgram, was one of the earliest proxy apps released to the HPC community. Within ExMatEx we use LULESH to explore programming models, including domain specific langiages, and as a representative coarse-scale component of our scale-bridging research and proxy app development.
The purpose of the ASPA proxy application is to enable the evaluation of a technique known as adaptive sampling on advanced computer architectures. Adaptive sampling is of interest in simulations involving multiple physical scales, wherein models of individual scales are combined using some form of scale bridging.