Engineering Papers⌕ Search

Engineering topics

Tomov, Stanimire

Publications and source records attributed to Tomov, Stanimire.

At least 19 records

MAGMA: Enabling exascale performance with accelerated BLAS and LAPACK for diverse GPU architectures

MAGMA (Matrix Algebra for GPU and Multicore Architectures) is a pivotal open-source library in the landscape of GPU-enabled dense and sparse linear algebra computations. With a repertoire of approximately 750 numerical routines across four precisions, MAGMA is deeply ingrained in the DOE software stack, playing a crucial role in high-performance computing. Notable projects such as ExaConstit, HiOP, MARBL, and STRUMPACK, among others, directly harness the capabilities of MAGMA. In addition, the MAGMA development team has been acknowledged multiple times for contributing to the vendors’ numerical software stacks. Looking back over the time of the Exascale Computing Project (ECP), we highlight how MAGMA has adapted to recent changes in modern HPC systems, especially the growing gap between CPU and GPU compute capabilities, as well as the introduction of low precision arithmetic in modern GPUs. We also describe MAGMA’s direct impact on several ECP projects. Maintaining portable performance across NVIDIA and AMD GPUs, and with current efforts toward supporting Intel GPUs, MAGMA ensures its adaptability and relevance in the ever-evolving landscape of GPU architectures.

97 MATHEMATICS AND COMPUTING↗

Port and optimize the CEED software stack to Aurora/Frontier EA (ECP Milestone Report)

The goal of this milestone was to port the CEED software stack, including Nek, MFEM and libCEED to the Frontier and Aurora early access hardware, work on optimizing the performance on AMD and Intel GPUs, and demonstrate impact in CEED-enabled ECP applications. As part of this milestone, we also performed a number of other activities, including continued optimizations for CEED applications at scale on Summit, performance evaluation on non-ECP hardware, including the Fugaku’s A64FX chip and the NVIDIA A100 GPU architecture, exploring the use of just-in-time compilations in applications at scale and potential of mixed precision optimizations in the CEED discretization and solver algorithms, and more. During the milestone period, the CEED team released new versions of 6 of its packages, including major releases of MFEM, NekRS and OCCA. We also organized the fifth CEED Annual meeting (CEED5AM) which included nearly 100 researchers from national labs, universities and industry.

97 MATHEMATICS AND COMPUTING↗

GPU algorithms for Efficient Exascale Discretizations

In this paper we describe the research and development activities in the Center for Efficient Exascale Discretization within the US Exascale Computing Project, targeting state-of-the-art high-order finite-element algorithms for high-order applications on GPU-accelerated platforms. Furthermore, we discuss the GPU developments in several components of the CEED software stack, including the libCEED, MAGMA, MFEM, libParanumal, and Nek projects. We report performance and capability improvements in several CEED-enabled applications on both NVIDIA and AMD GPU systems.

97 MATHEMATICS AND COMPUTING↗

Advances in Mixed Precision Algorithms: 2021 Edition

Over the last year, the ECP xSDK-multiprecision effort has made tremendous progress in developing and deploying new mixed precision technology and customizing the algorithms for the hardware deployed in the ECP flagship supercomputers. The effort also has succeeded in creating a cross-laboratory community of scientists interested in mixed precision technology and now working together in deploying this technology for ECP applications. In this report, we highlight some of the most promising and impactful achievements of the last year. Among the highlights we present are: Mixed precision IR using a dense LU factorization and achieving a 1.8× speedup on Spock; results and strategies for mixed precision IR using a sparse LU factorization; a mixed precision eigenvalue solver; Mixed Precision GMRES-IR being deployed in Trilinos, and achieving a speedup of 1.4× over standard GMRES; compressed Basis (CB) GMRES being deployed in Ginkgo and achieving an average 1.4× speedup over standard GMRES; preparing hypre for mixed precision execution; mixed precision sparse approximate inverse preconditioners achieving an average speedup of 1.2×; and detailed description of the memory accessor separating the arithmetic precision from the memory precision, and enabling memory-bound low precision BLAS 1/2 operations to increase the accuracy by using high precision in the computations without degrading the performance. We emphasize that many of the highlights presented here have also been submitted to peer-reviewed journals or established conferences, and are under peer-review or have already been published.

97 MATHEMATICS AND COMPUTING↗

Efficient exascale discretizations: High-order finite element methods

Efficient exploitation of exascale architectures requires rethinking of the numerical algorithms used in many large-scale applications. These architectures favor algorithms that expose ultra fine-grain parallelism and maximize the ratio of floating point operations to energy intensive data movement. One of the few viable approaches to achieve high efficiency in the area of PDE discretizations on unstructured grids is to use matrix-free/partially assembled high-order finite element methods, since these methods can increase the accuracy and/or lower the computational time due to reduced data motion. In this paper we provide an overview of the research and development activities in the Center for Efficient Exascale Discretizations (CEED), a co-design center in the Exascale Computing Project that is focused on the development of next-generation discretization software and algorithms to enable a wide range of finite element applications to run efficiently on future hardware. CEED is a research partnership involving more than 30 computational scientists from two US national labs and five universities, including members of the Nek5000, MFEM, MAGMA and PETSc projects. We discuss the CEED co-design activities based on targeted benchmarks, miniapps and discretization libraries and our work on performance optimizations for large-scale GPU architectures. We also provide a broad overview of research and development activities in areas such as unstructured adaptive mesh refinement algorithms, matrix-free linear solvers, high-order data visualization, and list examples of collaborations with several ECP and external applications.

97 MATHEMATICS AND COMPUTING↗

High-order algorithmic developments and optimizations for large-scale GPU-accelerated simulations (Milestone CEED-MS36)

The goal of this milestone was to improve the high-order software ecosystem for CEED-enabled ECP applications by making progress on efficient matrix-free kernels targeting forthcoming ECP architectures. These kernels included matrix-free preconditioning and the development of new set of CEED solver bake-off problems. As part of this milestone, we also released the next version of the CEED software stack, CEED-4.0, reported on results from several application collaborations, and documented the efforts of porting to AMD GPUs for Frontier and other modern architectures, such as Fugaku. The specific tasks addressed in this milestone were: (1) Port and run CEED benchmarks/miniapps on Frontier EA systems; (2) Demonstrate performant libCEED integration in MFEM, Nek and applications; (3) Matrix-free preconditioning of high-order operators; (4) Benchmark problems for fast high-order solvers on GPU platforms; and (5) Public release of CEED-4.0. The artifacts delivered include the next version of the CEED software stack, CEED-4.0, the next libCEED release, libCEED-0.8, and a number of developments integrated within applications to improve their GPU and CPU performance and capabilities. See the CEED website, https://ceed.exascaleproject.org and the CEED GitHub organization, https://github.com/ceed for more details.

97 MATHEMATICS AND COMPUTING↗

A survey of numerical linear algebra methods utilizing mixed-precision arithmetic

The efficient utilization of mixed-precision numerical linear algebra algorithms can offer attractive acceleration to scientific computing applications. Especially with the hardware integration of low-precision special-function units designed for machine learning applications, the traditional numerical algorithms community urgently needs to reconsider the floating point formats used in the distinct operations to efficiently leverage the available compute power. In this study, we provide a comprehensive survey of mixed-precision numerical linear algebra routines, including the underlying concepts, theoretical background, and experimental results for both dense and sparse linear algebra problems.

97 MATHEMATICS AND COMPUTING↗

Asynchronous Iterative Solvers for Extreme-Scale Computing

The Asynchronous Iterative Solvers for Extreme-Scale Computing (AsyncIS) project aims to explore more efficient numerical algorithms by decreasing their overhead. AsyncIS does this by replacing the outer Krylov subspace solver with an asynchronous optimized Schwarz method, thereby removing the global synchronization and bulk synchronous operations typically used in numerical codes. AsyncIS—a U.S. Department of Energy (DOE)-funded collaboration between Georgia Tech, the University of Tennessee, Knoxville, Temple University, and Sandia National Laboratories—also focuses on the development and optimization of asynchronous preconditioners (i.e., preconditioners that are generated and/or applied in an asynchronous fashion). The novel preconditioning algorithms that provide fine-grained parallelism enable preconditioned Krylov solvers to run efficiently on large-scale distributed systems and manycore accelerators like GPUs.

97 MATHEMATICS AND COMPUTING↗

Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems

Double-precision floating-point arithmetic (FP64) has been the de facto standard for engineering and scientific simulations for several decades. Problem complexity and the sheer volume of data coming from various instruments and sensors motivate researchers to mix and match various approaches to optimize compute resources, including different levels of floating-point precision. In recent years, machine learning has motivated hardware support for half-precision floating-point arithmetic. A primary challenge in high-performance computing is to leverage reduced-precision and mixed-precision hardware. We show how the FP16/FP32 Tensor Cores on NVIDIA GPUs can be exploited to accelerate the solution of linear systems of equations Ax = b without sacrificing numerical stability. The techniques we employ include multiprecision LU factorization, the preconditioned generalized minimal residual algorithm (GMRES), and scaling and auto-adaptive rounding to avoid overflow. We also show how to efficiently handle systems with multiple right-hand sides. On the NVIDIA Quadro GV100 (Volta) GPU, we achieve a 4×-5× performance increase and 5× better energy efficiency versus the standard FP64 implementation while maintaining an FP64 level of numerical stability.

GMRES↗

Support CEED-enabled ECP applications in their preparation for Aurora/Frontier

The goal of this milestone was to help CEED-enabled ECP applications (particularly ExaSMR, MARBL, ExaAM, ExaWind and E3SM) in their preparations for the Aurora and Frontier architectures. This work included collaboration with ECP vendors and porting and optimization of CEED’s benchmarks and miniapps to early access hardware. As part of this milestone, we also made the best bake-off problems and bake-off kernel implementation from Nek, MFEM, libParanumal and the external community available in the latest libCEED release, libCEED-0.7. During the milestone period we also organized, in virtual form, the fourth CEED Annual meeting (CEED4AM) which included representatives from ECP applications, vendors and software technology projects. The specific tasks addressed in this milestone were to: (1) Work with vendors to port and run CEED benchmarks on early access systems for Aurora and Frontier; (2) Make the best BP/BK implementations from Nek, MFEM, libParanumal and external community available in libCEED; (3) Organize the next CEED Annual meeting (CEED4AM); and (4) Optimize CEED applications and miniapps for Aurora and Frontier architectures. The artifacts delivered include the next libCEED release, libCEED-0.7, and a number of developments integrated within applications to improve their GPU performance and capabilities. See the CEED website, http://ceed.exascaleproject.org and the CEED GitHub organization, http://github.com/ceed for more details.

97 MATHEMATICS AND COMPUTING↗

CEED-MS34: Improve performance and capabilities of CEED-enabled ECP applications on Summit/Sierra. Exascale Computing Project Milestone Report

The main goal of this milestone was to help CEED-enabled ECP applications, including ExaSMR, MARBL, ExaWind and ExaAM, to improve their performance and capabilities on GPU systems like Summit and Lassen/Sierra. In addition, the CEED team also worked to: add and improve support for additional hardware and programming models in the CEED software components; release the next version of the CEED software stack, CEED-3.0; and demonstrate performance of libParanumal kernels in libCEED, Nek and MFEM. These additional tasks contributed directly to the main CEED-MS34 goal and will also play an important role in CEED’s future milestones.

97 MATHEMATICS AND COMPUTING↗

Performance tuning of CEED software and 1st and 2nd wave apps

The goal of this milestone was the performance tuning of the CEED software, as well as the use and tuning of CEED to accelerate the first and second wave of targeted ECP applications. In this milestone, the CEED team developed optimization techniques and tuned for performance the CEED software to accelerate the first and second wave target ECP applications. Specifically, the focus was on the following: (1) Efficient use of the memory sub-system for managing data motion and interactions among different physics packages and libraries; this included integration of the new developments in a libCEED 0.5 software release; (2) Enhancing the CEED libraries with GPU and AMD GPU support in close interaction with vendors; (3) Optimal data locality and motion, and enhanced scalability and parallelism; this included vendors interactions for improvements in data motion and making strong scaling easier and more efficient; and (4) Continue boosting performance for the first-wave and second-wave of ECP target applications, including ExaSMR, ExaWind, Urban, MARBL, E3SM, and ExaAM. A main part of this milestone was the performance tuning of the CEED first-wave and second-wave ECP applications. This included the ExaSMR application – Coupled Monte Carlo Neutronics and Fluid Flow Simulation of Small Modular Reactors (ORNL), the MARBL application – Next-Gen Multi-physics Simulation Code (LLNL), and the ExaAM ExaConstit – a miniapp for the Transforming Additive Manufacturing through Exascale Simulation applications project (ExaAM), as well as ExaWind, Urban, and E3SM. The artifacts delivered include performance improvements in CEED’s 1st and 2nd wave of applications, and tuned CEED software for various architectures through a number of backends, freely available in the CEED’s repository on GitHub. See the CEED website, http://ceed.exascaleproject.org and the CEED GitHub organization, http://github.com/ceed for more details. In addition to details and results from the above R&D efforts, in this document we are also reporting on other project-wide activities performed in Q4 of FY19 including: CEED’s third annual meeting, MFEM GPU kernels scaling improvements, libCEED 0.5 and NekRS releases, an overview paper covering latest results on a set of bake-off high-order finite element problems (BPs), and other outreach efforts.

97 MATHEMATICS AND COMPUTING↗

Optimizing the Fast Fourier Transform Using Mixed Precision on Tensor Core Hardware

The Fast Fourier Transform is a fundamental tool in scientific and technical computation. The highly parallelizable nature of the algorithm makes it a suitable candidate for GPU acceleration. This paper focuses on exploiting the speedup due to using the half precision multiplication capability of the latest GPUs' tensor core hardware without significantly degrading the precision of the Fourier Transform result. We develop an algorithm that dynamically splits the input single precision dataset into two half precision sets at the lowest level, uses half precision multiplication, and recombines the result at a later step. This work paves the way for using tensor cores for high precision inputs.

Sorna, Anumeena↗

Performance tuning of CEED software and first wave apps

The goal of this milestone was the performance tuning of the CEED software and first wave apps. In this milestone, the CEED team developed optimization techniques and tuned for performance the CEED software. Specifically, the focus was on the following: (1) Fast finite element operator storage and evaluation using partial assembly/matrix-free algorithms that take advantage of tensor-product element structure; (2) Architecture optimizations, including the development of performant discretization libraries targeting heterogeneous systems with multi-core CPUs, GPUs, and/or many-core CPUs in single node, cluster, or large-scale ECP supercomputer configurations; and (3) Global kernels for finite element operators, including the development of algorithms and libraries for fast gather-scatter exchanges for action assembly on unstructured graphs and for components needed for global linear solvers. A main part of this milestone was the performance tuning of the CEED first-wave ECP applications. This included the ExaSMR application – Coupled Monte Carlo Neutronics and Fluid Flow Simulation of Small Modular Reactors (ORNL), and the MARBL application – Next-Gen Multi-physics Simulation Code (LLNL). The artifacts delivered include performance improvements in CEED’s 1st wave of applications, and tuned CEED software for various architectures through a number of backends, freely available in the CEED’s repository on GitHub. See the CEED website, http://ceed.exascaleproject.org and the CEED GitHub organization, http://github.com/ceed for more details. In addition to details and results from the above R&D efforts, in this document we are also reporting on other project-wide activities performed in Q4 of FY18 including: CEED’s second annual meeting, a successful minisymposium at the premier international conference on high-order methods (ICOSAHOM), making the CEED milestone reports publicly available, new Laghos release, and other outreach efforts.

97 MATHEMATICS AND COMPUTING↗

Computational Benefit of GPU Optimization for the Atmospheric Chemistry Modeling

Abstract Global chemistry‐climate models are computationally burdened as the chemical mechanisms become more complex and realistic. Optimization for graphics processing units (GPU) may make longer global simulation with regional detail possible, but limited study has been done to explore the potential benefit for the atmospheric chemistry modeling. Hence, in this study, the second‐order Rosenbrock solver of the chemistry module of CAM4‐Chem is ported to the GPU to gauge potential speed‐up. We find that on the CPU, the fastest performance is achieved using the Intel compiler with a block interleaved memory layout. Different combinations of compiler and memory layout lead to ~11.02× difference in the computational time. In contrast, the GPU version performs the best when using a combination of fully interleaved memory layout with block size equal to the warp size, CUDA streams for independent kernels, and constant memory. Moreover, the most efficient data transfer between CPU and GPU is gained by allocating the memory contiguously during the data initialization on the GPU. Compared to one CPU core, the speed‐up of using one GPU alone reaches a factor of ~11.7× for the computation alone and ~3.82× when the data transfer between CPU and GPU is considered. Using one GPU alone is also generally faster than the multithreaded implementation for 16 CPU cores in a compute node and the single‐source solution (OpenACC). The best performance is achieved by the implementation of the hybrid CPU/GPU version, but rescheduling the workload among the CPU cores is required before the practical CAM4‐Chem simulation.

54 ENVIRONMENTAL SCIENCES↗