Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Distributed Memory Parallel”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 271 records · Page 15

Computing for Finance

The finance sector is one of the driving forces for the use of distributed or Grid computing for business purposes. The speakers will review the state-of-the-art of high performance computing in the financial sector, and provide insight into how different types of Grid computing – from local clusters to global networks - are being applied to financial applications. They will also describe the use of software and techniques from physics, such as Monte Carlo simulations, in the financial world. There will be four talks of 20min each. The talk abstracts and speaker bios are listed below. This will be followed by a Q&A; panel session with the speakers. From 19:00 onwards there will be a networking cocktail for audience and speakers. This is an EGEE / CERN openlab event organized in collaboration with the regional business network rezonance.ch. A webcast of the event will be made available for subsequent viewing, along with powerpoint material presented by the speakers. Attendance is free and open to all. Registration is mandatory via www.rezonance.ch, including for CERN staff. 1. Overview of High Performance Computing in the Financial Industry Michael Yoo, Managing Director, Head of the Technical Council, UBS Presentation will describe the key business challenges driving the need for HPC solutions, describe the means in which those challenges are being addressed within UBS (such as GRID) as well as the limitations of some of these solutions, and assess some of the newer HPC technologies which may also play a role in the Financial Industry in the future. Speaker Bio: Michael originally joined the former Swiss Bank Corporation in 1994 in New York as a developer on a large data warehouse project. In 1996 he left SBC and took a role with Fidelity Investments in Boston. Unable to stay away for long, he returned to SBC in 1997 while working for Perot Systems in Singapore. Finally, in 1998 he formally returned to UBS in Stamford following the merger with SBC and has remained with UBS for the past 9 years. During his tenure at UBS, he has had a number of leadership roles within IT in development, support and architecture. In 2006 Michael relocated to Switzerland to take up his current role as head of the UBS IB Technical Council, responsible for the overall technology strategy and vision of the Investment Bank. One of Michael's key responsibilities is to manage the UBS High Performance Computing Research Lab and he has been involved in a number of initiatives in the HPC space. 2. Grid in the Commercial WorldFred Gedling, Chief Technology Officer EMEA and Senior Vice President Global Services, DataSynapse Grid computing gets mentions in the press for community programs starting last decade with Government, national and supranational initiatives in grid receive some press. One of the IT-industries' best-kept secrets is the use of grid computing by commercial organizations with spectacular results. Grid Computing and its evolution into Application Virtualization is discussed and how this is key to the next generation data center. Speaker Bio: Fred Gedling holds the joint roles of Chief Technology Officer for EMEA and Senior Vice President of Global Services at DataSynapse, a global provider of application virtualisation software. Based in London and working closely with organisations seeking to optimise their IT infrastructures, Fred offers unique insights into the technology of virtualisation as well as the methodology of establishing ROI and rapid deployment to the immediate advantage of the business. Fred has more than fifteen years experience of enterprise middleware and high-performance infrastructures. Prior to DataSynapse he worked in high performance CRM middleware and was the CTO EMEA for New Era of Networks (NEON) during the rapid growth of Enterprise Application Integration. His 25-year career in technology also includes management positions at Goldman Sachs and Stratus Computer. Fred holds a First Class Bsc (Hons) degree in Physics with Astrophysics from the University of Leeds and had the privilege of being a summer student at CERN.3. Opportunities for gLite in finance and related industries Adam Vile, Head of Grid, HPC and Technical Computing, Excelian Ltd.gLite, the Grid software developed by the EGEE project, has been exceedingly successful as an enabling infrastructure, and has been a massive success in bringing together scientific and technical communities to provide the compute power to address previously incomputable problems. Not so in the finance industry. In its current form gLite would be a business disabler. There are other middleware tools that solve the finance communities compute problems much better. Things are moving on, however. There are moves afoot in the open source community to evolve the technology to address other, more sophisticated needs such as utility and interactive computing. In this talk, I will describe how Excelian is providing Grid consultancy services for the finance community and how, through its relationship to the EGEE project, Excelian is helping to identify and exploit opportunities as the research and business worlds converge. Because of the strong third party presence in the finance industry, such opportunities are few and far between, but they are there, especially as we expand sideways into related verticals such as the smaller hedge funds and energy companies. This talk will give an overview of the barriers to adoption of gLite in the finance industry and highlight some of the opportunities offered in this and related industries as the ideas around Grid mature. Speaker Bio: Dr Adam Vile is a senior consultant and head of the Grid and HPC practice at Excelian, a consultancy that focuses on financial markets professional services. He has spent many years in investment banking, as a developer, project manager and architect in both front and back office. Before joining Excelian he was senior Grid and HPC architect at Barclays Capital. Prior to joining investment banking, Adam spent a number of years lecturing in IT and mathematics at a UK University and maintains links with academia through lectures, research and through validation and steering of postgraduate courses. He is a chartered mathematician and was the conference chair of the Institute of Mathematics and its Applications first conference in computational Finance.4. From Monte Carlo to Wall Street Daniel Egloff, Head of Financial Engineering Computing Unit, Zürich Cantonal Bank High performance computing techniques provide new means to solve computationally hard problems in the financial service industry. First I consider Monte Carlo simulation and illustrate how it can be used to implement a sophisticated credit risk management and economic capital framework. From a HPC perspective, basic Monte Carlo simulation is embarrassingly parallel and can be implemented efficiently on distributed memory clusters. Additional difficulties arise for adaptive variance reduction schemes, if the information content in a sample is very small, and if the amount of simulated date becomes huge such that incremental processing algorithms are indispensable. We discuss the business value of an advanced credit risk quantification which is particularly compelling in these days. While Monte Carlo simulation is a very versatile tool it is not always the preferred solution for the pricing of complex products like multi asset options, structured products, or credit derivatives. As a second application I show how operator methods can be used to develop a pricing framework. The scalability of operator methods relies heavily on optimized dense matrix-matrix multiplications and requires specialized BLAS level-3 implementations provided by specialized FPGA or GPU boards. Speaker Bio: Daniel Egloff studied mathematics, theoretical physics, and computer science at the University of Zurich and the ETH Zurich. He holds a PhD in Mathematics from University of Fribourg, Switzerland. After his PhD he started to work for a large Swiss insurance company in the area of asset and liability management. He continued his professional career in the consulting industry. At KPMG and Arthur Andersen he consulted international clients and implemented quantitative risk management solutions for financial institutions and insurance companies. In 2002 he joined Zurich Cantonal Bank. He was assigned to develop and implement credit portfolio risk and economic capital methodologies. He built up a competence center for high performance and cluster computing. Currently, Daniel Egloff is heading the Financial Computing unit in the ZKB Financial Engineering division. He and his team is engineering and operating high performance cluster applications for computationally intensive problems in financial risk management.

Source record↗

The New MCNP6 Depletion Capability

The first MCNP based inline Monte Carlo depletion capability was officially released from the Radiation Safety Information and Computational Center as MCNPX 2.6.0. Both the MCNP5 and MCNPX codes have historically provided a successful combinatorial geometry based, continuous energy, Monte Carlo radiation transport solution for advanced reactor modeling and simulation. However, due to separate development pathways, useful simulation capabilities were dispersed between both codes and not unified in a single technology. MCNP6, the next evolution in the MCNP suite of codes, now combines the capability of both simulation tools, as well as providing new advanced technology, in a single radiation transport code. We describe here the new capabilities of the MCNP6 depletion code dating from the official RSICC release MCNPX 2.6.0, reported previously, to the now current state of MCNP6. NEA/OECD benchmark results are also reported. The MCNP6 depletion capability enhancements beyond MCNPX 2.6.0 reported here include: (1) new performance enhancing parallel architecture that implements both shared and distributed memory constructs; (2) enhanced memory management that maximizes calculation fidelity; and (3) improved burnup physics for better nuclide prediction. MCNP6 depletion enables complete, relatively easy-to-use depletion calculations in a single Monte Carlo code. The enhancements described here help provide a powerful capability as well as dictate a path forward for future development to improve the usefulness of the technology.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

GeoPhysical Analysis Code

GPAC is a code that integrates open source libraries for element formulations, linear algebra, and I/O with two main LLNL-Written components: (i) a set of standard finite elements physics solvers for rersolving Darcy fluid flow, explicit mechanics, implicit mechanics, and fluid-mediated fracturing, including resolution of contact both implicity and explicity, and (ii) a MPI-based parallelization implementation for use on generic HPC distributed memory architectures. The resultant code can be used alone for linearly elastic problems and problems involving hydraulic fracturing, where the mesh topology is dynamically changed. The key application domain is for low-rate stimulation and fracture control in subsurface reservoirs (e.g., enhanced geothermal sites and unconventional shale gas stimulation). GPAC also has interfaces to call external libraries for, e.g., material models and equations of state; however, LLNL-developed EOS and material models will not be part of the current release.

Source record↗

GeoPhysical Analysis Code

GPAC is a code that integrates open source libraries for element formulations, linear algebra, and I/O with two main LLNL-written components: (i) a set of standard finite, discrete, and discontinuous displacement element physics solvers for resolving Darcy fluid flow, explicit mechanics, implicit mechanics, fault rupture and earthquake nucleation, and fluid-mediated fracturing, including resolution of physcial behaviors both implicity and explicity, and (ii) a MPI-based parallelization implementation for use on generic HPC distributed memory architectures. The resultant code can be used alone for linearly elastic problems; ploblems involving hydraulic fracturing, where the mesh topology is dynamically changed; fault rupture modeling and seismic risk assessment; and general granular materials behavior. The key application domain is for low-rate stimulation and fracture control in subsurface reservoirs (e.g., enhanced geothermal sites and unconventional shale gas stimulation). GPAC also has interfaces to call external libraries for , e.g., material models and equations of state; however, LLNL-developed EOS and material models will not be part of the current release. CPAC's secondary applications include modeling fault evolution for predicting the statistical distribution of earthquake events and to capture granular materials behavior under different load paths.

Source record↗

The new MCNP6 depletion capability

The first MCNP based in-line Monte Carlo depletion capability was officially released from the Radiation Safety Information and Computational Center as MCNPX 2.6.0. Both the MCNP5 and MCNPX codes have historically provided a successful combinatorial geometry based, continuous energy, Monte Carlo radiation transport solution for advanced reactor modeling and simulation. However, due to separate development pathways, useful simulation capabilities were dispersed between both codes and not unified in a single technology. MCNP6, the next evolution in the MCNP suite of codes, now combines the capability of both simulation tools, as well as providing new advanced technology, in a single radiation transport code. We describe here the new capabilities of the MCNP6 depletion code dating from the official RSICC release MCNPX 2.6.0, reported previously, to the now current state of MCNP6. NEA/OECD benchmark results are also reported. The MCNP6 depletion capability enhancements beyond MCNPX 2.6.0 reported here include: (1) new performance enhancing parallel architecture that implements both shared and distributed memory constructs; (2) enhanced memory management that maximizes calculation fidelity; and (3) improved burnup physics for better nuclide prediction. MCNP6 depletion enables complete, relatively easy-to-use depletion calculations in a single Monte Carlo code. The enhancements described here help provide a powerful capability as well as dictate a path forward for future development to improve the usefulness of the technology. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Distributed non-equilibrium Green’s function algorithms for the simulation of nanoelectronic devices with scattering

Through the non-equilibrium Green’s function (NEGF) formalism, quantum-scale device simulation can be performed with the inclusion of electron-phonon scattering. Yet, the simulation of realistically sized devices under the NEGF formalism typically requires prohibitive amounts of memory and computation time. Two of the most demanding computational problems for NEGF simulation involve mathematical operations with structured matrices called semiseparable matrices. In this work, we present parallel approaches for these computational problems which allow for efficient distribution of both memory and computation based upon the underlying device structure. This is vital when simulating realistically sized devices due to the aforementioned computational burdens. First, we consider determining a distributed compact representation for the retarded Green’s function matrix G R . This compact representation is exact and allows for any entry in the matrix to be generated through the inherent semiseparable structure. The second parallel operation allows for the computation of electron density and current characteristics for the device. Specifically, matrix products between the distributed representation for the semiseparable matrix G R and the self-energy scattering terms in Σ < produce the less-than Green’s function G < . As an illustration of the computational efficiency of our approach, we stably generate the mobility for nanowires with cross-sectional sizes as large as 4.5 nm, assuming an atomistic model with scattering.

47 OTHER INSTRUMENTATION↗

Enabling the High Level Synthesis of Data Analytics Accelerators

Conventional High Level Synthesis (HLS) tools mainly tar- get compute intensive kernels typical of digital signal pro- cessing applications. We are developing techniques and ar- chitectural templates to enable HLS of data analytics appli- cations. These applications are memory intensive, present fine-grained, unpredictable data accesses, and irregular, dy- namic task parallelism. We discuss an architectural tem- plate based around a distributed controller to efficiently ex- ploit thread level parallelism. We present a memory in- terface that supports parallel memory subsystems and en- ables implementing atomic memory operations. We intro- duce a dynamic task scheduling approach to efficiently ex- ecute heavily unbalanced workload. The templates are val- idated by synthesizing queries from the Lehigh University Benchmark (LUBM), a well know SPARQL benchmark.

Data Analytics↗

Parallel sorting algorithm classification: is manual instrumentation necessary?

Understanding parallel algorithms is crucial for accelerating scientific simulations on complex, distributed memory, high-performance computers. Modern algorithm classification approaches learn semantics directly from source code to differentiate between algorithms, however, accessing source code is not always possible. We can learn about parallel algorithms from observing their performance, as programs running the same algorithms and using the same hardware should exhibit similar performance characteristics. We present an approach to learn algorithm classes from parallel performance data directly in order to classify algorithms without access to the source code. We extend previous work to enable classifying parallel sorting algorithms using automatic instrumentation instead of requiring manual region annotations in the source code. In this work, we design and demonstrate a study for classification of parallel sorting algorithms using parallel performance data collected from automatic instrumentation, and evaluate the performance of our new methodology on classification. We leverage Caliper to collect the performance data, Thicket for our exploratory data analysis (EDA), and PyTorch and Scikit-learn to evaluate the effectiveness of random forests, support vector machines (SVMs), decision trees, neural networks, and logistic regressions on parallel performance data. Additionally, we study noise in parallel performance data, whether the removal of noise and pre-processing of the data is necessary to accurately classify parallel sorting algorithms, and determine the effectiveness of features created from performance data. In conclusion, we demonstrate classification accuracy for these five different models of up to 97.7% across four different parallel algorithm classes.

Algorithm Classification↗

New Results on Communication- and Memory-Aware Load Balancing Model and Algorithms

While load balancing in distributed-memory computing has been well-studied, we present an innovative approach to this problem: a unified, reduced-order model that combines three key components to describe “work” in a distributed system: computation, communication, and memory. Our model enables an optimizer to explore complex tradeoffs in task placement, such as augmented parallelism, at the expense of data replication increasing memory usage. We propose a fully distributed, heuristic-based load balancing optimization algorithm, and demonstrate that it quickly finds close-to-optimal solutions. We formalize the complex optimization problem as a mixed-integer linear program, and compare it to our strategy. Finally, we show that when applied to an electromagnetics code, our approach obtains up to 2.3x speedups for the imbalanced execution.

97 MATHEMATICS AND COMPUTING↗

Hierarchical computation in the SPMD programming model

Large-scale parallel machines are programmed mainly with the single program, multiple data (SPMD) model of parallelism. While this model has advantages of scalability and simplicity, it does not fit well with divide-and-conquer parallelism or hierarchical machines that mix shared and distributed memory. In this paper, we define the recursive single program, multiple data model (RSPMD) that extends SPMD with a hierarchical team mechanism to support hierarchical algorithms and machines. We implement this model in the Titanium language and describe how to eliminate a class of deadlocks by ensuring alignment of collective operations. We present application case studies evaluating the RSPMD model, showing that it enables divide-and-conquer algorithms such as sorting to be elegantly expressed and that team collective operations increase performance of conjugate gradient by up to a factor of two. The model also facilitates optimizations for hierarchical machines, improving scalability of particle in cell by 8x and performance of sorting and a stencil code by up to 40% and 14%, respectively.

Kamil, A↗

Distributed out-of-memory NMF on CPU/GPU architectures

We propose an efficient distributed out-of-memory implementation of the non-negative matrix factorization (NMF) algorithm for heterogeneous high-performance-computing systems. The proposed implementation is based on prior work on NMFk, which can perform automatic model selection and extract latent variables and patterns from data. In this work, we extend NMFk by adding support for dense and sparse matrix operation on multi-node, multi-GPU systems. The resulting algorithm is optimized for out-of-memory problems where the memory required to factorize a given matrix is greater than the available GPU memory. Memory complexity is reduced by batching/tiling strategies, and sparse and dense matrix operations are significantly accelerated with GPU cores (or tensor cores when available). Input/output latency associated with batch copies between host and device is hidden using CUDA streams to overlap data transfers and compute asynchronously, and latency associated with collective communications (both intra-node and inter-node) is reduced using optimized NVIDIA Collective Communication Library (NCCL) based communicators. Benchmark results show significant improvement, from 32X to 76x speedup, with the new implementation using GPUs over the CPU-based NMFk. Good weak scaling was demonstrated on up to 4096 multi-GPU cluster nodes with approximately 25,000 GPUs when decomposing a dense 340 Terabyte-size matrix and an 11 Exabyte-size sparse matrix of density 10 -6 .

97 MATHEMATICS AND COMPUTING↗

Sparse, iterative simulation methods for one-dimensional laminar flames

Sparse, iterative simulation methods for one-dimensional laminar flames are proposed. The resulting steady and unsteady flame solvers exploit approximate Jacobians to greatly reduce the computational cost associated with matrix operations. The constant, non-unity Lewis number assumption is introduced to further reduce the computational cost. The solvers are also parallelized to reduce time-to-solution on distributed memory computer systems. Computed laminar flame speeds and species profiles for a range of chemical mechanisms (from 10 to 2878 species) are compared against a well-validated commercial code and are found to be consistent within solver tolerances. The computation times of both the unsteady and steady solutions increase only linearly with the number of species, which is a significant improvement over the quadratic or cubic scaling of existing steady-state flame solvers. For the largest mechanism tested, the steady-state flame solver is two orders of magnitude faster than commonly-used codes. The use of an approximate Jacobian is shown to reduce the rate of convergence for the steady-state solver, but does not significantly affect the domain of convergence. The steady-state solver with approximate Jacobian is thus well suited for computationally efficient laminar flame speed sweeps with large kinetic mechanisms.

42 ENGINEERING↗

On the use of flux limiters in the discrete ordinates method for 3D radiation calculations in absorbing and scattering media

The application of flux limiters to the discrete ordinates method (DOM), S{sub N}, for radiative transfer calculations is discussed and analyzed for 3D enclosures for cases in which the intensities are strongly coupled to each other such as: radiative equilibrium and scattering media. A Newton-Krylov iterative method (GMRES) solves the final systems of linear equations along with a domain decomposition strategy for parallel computation using message passing libraries in a distributed memory system. Ray effects due to angular discretization and errors due to domain decomposition are minimized until small variations are introduced by these effects in order to focus on the influence of flux limiters on errors due to spatial discretization, known as numerical diffusion, smearing or false scattering. Results are presented for the DOM-integrated quantities such as heat flux, irradiation and emission. A variety of flux limiters are compared to 'exact' solutions available in the literature, such as the integral solution of the RTE for pure absorbing-emitting media and isotropic scattering cases and a Monte Carlo solution for a forward scattering case. Additionally, a non-homogeneous 3D enclosure is included to extend the use of flux limiters to more practical cases. The overall balance of convergence, accuracy, speed and stability using flux limiters is shown to be superior compared to step schemes for any test case.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Direct pore-level modeling of incompressible fluid flow in porous media

We present a dynamic particle-based model for direct pore-level modeling of incompressible viscous fluid flow in disordered porous media. The model is capable of simulating flow directly in three-dimensional high-resolution micro-CT images of rock samples. It is based on moving particle semi-implicit (MPS) method. We modify this technique in order to improve its stability for flow in porous media problems. Using the micro-CT image of a rock sample, the entire medium, i.e., solid and fluid, is discretized into particles. The incompressible Navier-Stokes equations are then solved for each particle using the MPS summations. The model handles highly irregular fluid-solid boundaries effectively. An algorithm to split and merge fluid particles is also introduced. To handle the computational load, we present a parallel version of the model that runs on distributed memory computer clusters. The accuracy of the model is validated against the analytical, numerical, and experimental data available in the literature. The validated model is then used to simulate both unsteady- and steady-state flow of an incompressible fluid directly in a representative elementary volume (REV) size micro-CT image of a naturally-occurring sandstone with 3.398 {mu}m resolution. We analyze the quality and consistency of the predicted flow behavior and calculate absolute permeability using the steady-state flow rate.

97 MATHEMATICS AND COMPUTING↗

HipMCL: a high-performance parallel implementation of the Markov clustering algorithm for large-scale networks

Biological networks capture structural or functional properties of relevant entities such as molecules, proteins or genes. Characteristic examples are gene expression networks or protein–protein interaction networks, which hold information about functional affinities or structural similarities. Such networks have been expanding in size due to increasing scale and abundance of biological data. While various clustering algorithms have been proposed to find highly connected regions, Markov Clustering (MCL) has been one of the most successful approaches to cluster sequence similarity or expression networks. Despite its popularity, MCL’s scalability to cluster large datasets still remains a bottleneck due to high running times and memory demands. In this paper, we present High-performance MCL (HipMCL), a parallel implementation of the original MCL algorithm that can run on distributed-memory computers. We show that HipMCL can efficiently utilize 2000 compute nodes and cluster a network of ~70 million nodes with ~68 billion edges in ~2.4 h. By exploiting distributed-memory environments, HipMCL clusters large-scale networks several orders of magnitude faster than MCL and enables clustering of even bigger networks. Finally, HipMCL is based on MPI and OpenMP and is freely available under a modified BSD license.

59 BASIC BIOLOGICAL SCIENCES↗

Massively parallel direct numerical simulations of forced compressible turbulence: a hybrid MPI/OpenMP approach. In:XSEDE '12 Proceedings of the 1st Conference of the Extreme Science and Engineering Discovery Environment: Bridging from the eXtreme to the campus and beyond, Article No. 23

A highly scalable simulation code for turbulent flows which solves the fully compressible Navier-Stokes equations is presented. The code, which supports one, two and three dimensional domain decompositions is shown to scale well on up to 262,144 cores. Introducing multiple levels of parallelism based on distributed message passing and shared-memory paradigms results in a reduction of up to 33% of communication time at large core counts. The code has been used to generate a large database of homogeneous isotropic turbulence in a stationary state created by forcing the largest scales in the flow. The scaling of spectra of velocity and density fluctuations are presented. While the former follow classical theories strictly valid for incompressible flows, the latter presents a more complicated behavior. Fluctuations in velocity gradients and derived quantities exhibit extreme though rare fluctuations, a phenomenon known as intermittency. The simulations presented provide data to disentangle Reynolds and Mach number effects.

Jagannathan, Shriram↗

Implementing Molecular Dynamics for Hybrid High Performance Computers - 1. Short Range Forces

The use of accelerators such as general-purpose graphics processing units (GPGPUs) have become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power requirements. Hybrid high performance computers, machines with more than one type of floating-point processor, are now becoming more prevalent due to these advantages. In this work, we discuss several important issues in porting a large molecular dynamics code for use on parallel hybrid machines - 1) choosing a hybrid parallel decomposition that works on central processing units (CPUs) with distributed memory and accelerator cores with shared memory, 2) minimizing the amount of code that must be ported for efficient acceleration, 3) utilizing the available processing power from both many-core CPUs and accelerators, and 4) choosing a programming model for acceleration. We present our solution to each of these issues for short-range force calculation in the molecular dynamics package LAMMPS. We describe algorithms for efficient short range force calculation on hybrid high performance machines. We describe a new approach for dynamic load balancing of work between CPU and accelerator cores. We describe the Geryon library that allows a single code to compile with both CUDA and OpenCL for use on a variety of accelerators. Finally, we present results on a parallel test cluster containing 32 Fermi GPGPUs and 180 CPU cores.

43 PARTICLE ACCELERATORS↗

Direct numerical simulations for hybrid rocket boundary layers: Performance modeling and scaling

This paper presents a comprehensive performance and scaling analysis of direct numerical simulations for reacting boundary layers, focusing on slab burner configurations. Using a PETSc-based finite volume CFD framework, the study evaluates the scalability and computational cost of flow, chemistry, and radiation evaluations across 2D and 3D simulations. Polymethyl methacrylate (PMMA) is the fuel with pure O 2 as the oxidizer, modeled using a detailed chemical kinetics mechanism with 113 species and 660 reactions. A ray-tracing-based radiation solver, designed for distributed memory applications, is implemented to model radiation heat transfer. Parallel scalability is analyzed for the coupled flow, chemistry, and radiation heat transfer processes. Weak and strong scaling studies are conducted on up to 15,000 computational ranks, revealing robust performance when flow cells exceed 200 per rank. Chemistry evaluations dominate the computational cost in large 3D simulations, accounting for approximately 40% of the total runtime, while flow processes contribute around 35%, and radiation solver contributions remain below 10% due to reduced evaluation frequencies. GPU accelerated chemistry evaluation, implemented with Zero-RK, demonstrates significant promise, achieving up to a 4x speedup for workloads exceeding 30,000 cells per GPU. However, diminishing returns are observed for smaller workloads due to CPU-GPU communication overhead. This study identifies key challenges, including memory bottlenecks and the effects of domain partitioning on flow scalability, while highlighting the potential of GPU-accelerated chemistry to reduce computational costs. In conclusion, these findings provide realizable run configurations for 2D, 3D, and GPU-accelerated cases, offering insights for optimizing reactive flow solvers.

CFD Scalability↗