Search NASA⌕ Search

Engineering topics

Prokopenko, Andrey

Publications and source records attributed to Prokopenko, Andrey.

18 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Advances in ArborX to support exascale applications

ArborX is a performance portable geometric search library developed as part of the Exascale Computing Project (ECP). In this paper, we explore a collaboration between ArborX and a cosmological simulation code HACC. Large cosmological simulations on exascale platforms encounter a bottleneck due to the in-situ analysis requirements of halo finding, a problem of identifying dense clusters of dark matter (halos). This problem is solved by using a density-based DBSCAN clustering algorithm. With each MPI rank handling hundreds of millions of particles, it is imperative for the DBSCAN implementation to be efficient. In addition, the requirement to support exascale supercomputers from different vendors necessitates performance portability of the algorithm. We describe how this challenge problem guided ArborX development, and enhanced the performance and the scope of the library. We explore the improvements in the basic algorithms for the underlying search index to improve the performance, and describe several implementations of DBSCAN in ArborX. Further, we report the history of the changes in ArborX and their effect on the time to solve a representative benchmark problem, as well as demonstrate the real world impact on production end-to-end cosmology simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

PANDORA: A Parallel Dendrogram Construction Algorithm for Single Linkage Clustering on GPU

This paper introduces Pandora, a parallel algorithm for computing dendrograms, the hierarchical cluster trees for single linkage clustering (SLC). Current parallel approaches construct dendrograms by partitioning a minimum spanning tree and removing edges. However, they struggle with skewed, hard-to-parallelize real-world dendrograms. Consequently, computing dendrograms is the sequential bottleneck in HDBSCAN*[21], a popular SLC variant. Pandora uses recursive tree contraction to address this limitation. Pandora contracts nodes to construct progressively smaller trees. It computes the smallest contracted dendrogram and expands it by inserting contracted edges. This recursive strategy is highly parallel, skew-independent, work-optimal, and well-suited for GPUs and multicores. We develop a performance portable implementation of Pandora in Kokkos[31] and evaluate its performance on multicore CPUs and multi-vendor GPUs (e.g., Nvidia, AMD) for dendrogram construction in HDBSCAN*. Multithreaded Pandora is 2.2x faster than the current best-multithreaded implementation. Our GPU version achieves 6-20x speedup on AMD GPUs and 10-37x on NVIDIA GPUs over multithreaded Pandora. Pandora removes HDBSCAN*’s sequential bottleneck, greatly boosting efficiency, particularly with GPUs.

Sao, Piyush↗
DOE OSTI2024

Revising Apetrei’s bounding volume hierarchy construction algorithm to allow stackless traversal

Stackless traversal is a technique to speed up range queries by avoiding usage of a stack during the tree traversal. One way to achieve that is to transform a given binary tree to store a left child and a skip-connection (also called an escape index). In general, this operation requires an additional tree traversal during the tree construction. For some tree structures, however, it is possible to achieve the same result at a reduced cost. We propose one such algorithm for a GPU hierarchy construction algorithm proposed by Karras in Karras 2012. Furthermore, we show that our algorithm also works with the improved algorithm proposed by Apetrei in Apetrei 2014, despite a different ordering of the internal nodes. We achieve that by modifying Apetrei’s algorithm to restore the original Karras’ ordering of the internal nodes. Using the modified algorithm, we show how to construct a hierarchy suitable for a stackless traversal in a single bottom-up pass.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

An implicit barotropic mode solver for MPAS-ocean using a modern Fortran solver interface

Here, we demonstrate use of a modern Fortran solver interface to manage solver algorithms for an implicit barotropic mode solver in the Model for Predictions Across Scales-Ocean (MPAS-O). ForTrilinos, a Fortran interface to Trilinos that contains a large collection of solver capabilities written in C++, has been implemented in MPAS-O to provide access to a suite of linear solver options. By virtue of the simplified wrapper and interface generator (SWIG) automation tool that generates modern Fortran interfaces to C++ code, we were able to implement the Fortran solver interface in MPAS-O using a familiar Fortran coding style while minimizing performance degradation. The ForTrilinos solver interface is written within MPAS-O’s time stepping modules as a subroutine in conjunction with MPAS-O code. Applied to an idealized ocean and a high-resolution realistic ocean test case, parallel performance of ForTrilinos solvers is examined. It is found that parallel scalability of the ForTrilinos solvers is highly dependent on the number of global synchronization points per solver iteration in each iterative solver algorithm. ForTrilinos solvers perform best compared to the Fortran hand-crafted (FHC) solver when the amount of work per processor is large enough. However, parallel scalability is better with the FHC solver and so when the work per core is modest FHC outperforms ForTrilinos. The intercomparison between the ForTrilinos and FHC solvers reveals that this performance hit in the ForTrilinos solver mostly comes from the global synchronization process, while suggesting that the matrix-vector multiplication process in the FHC solver needs to be optimized for better performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Latency and Bandwidth Microbenchmarks of US Department of Energy Systems in the June 2023 Top 500 List

As a rule, Top 500 class supercomputers are extensively benchmarked as part of their acceptance testing process. However, barring publicly posted LINPACK / HPCG results, most benchmark results are often inaccessible outside the hosting institution. Moreover, these higher level benchmarks do not provide easy answers to common questions such as “What is the realizable memory bandwidth?” or “What is the launch latency on the accelerator?” To partially address these issues, we executed selected single-node micro-benchmarks — focused on latencies and memory bandwidth — on every US Department of Energy system above rank 150 of the June 2023 Top 500 list, excepting NERSC’s Cori and ORNL’s Frontier TDS (now decommissioned or repurposed). We hope to provide an easy “first stop” reference for users of current Top 500 systems and inspire users and administrators of other Top 500 systems to similarly compile and make available benchmark results for their systems.

Siefert, Christopher M.↗
DOE OSTI2023

Fast tree-based algorithms for DBSCAN for low-dimensional data on GPUs

DBSCAN is a well-known density-based clustering algorithm to discover arbitrary shape clusters. While conceptually simple in serial, the algorithm is challenging to efficiently parallelize on manycore GPU architectures. Common pitfalls, such as asynchronous range query calls, result in high thread execution divergence in many implementations. In this paper, we propose a new framework for GPU-accelerated DBSCAN, and describe two tree-based algorithms within that framework. Both algorithms fuse the search for neighbors with updating cluster information, but differ in their treatment of dense regions of the data. We show that the time taken to compute clusters is at most twice that of determination of the neighbors. We compare the proposed algorithms with existing CPU and GPU implementations, and demonstrate their competitiveness and performance using a fast traversal structure (bounding volume hierarchy) for low dimensional data. We also show that the memory usage can be reduced by processing object neighbors dynamically without storing them.

Prokopenko, Andrey↗
DOE OSTI2023

ExaWind: Then and Now

The scientific goal of the ExaWind project is to advance our fundamental understanding of the flow physics governing whole wind plant performance, including wake formation, complex terrain impacts, and turbine-turbine-interaction effects. The primary application codes in the ExaWind environment are Nalu-Wind, an unstructured-grid computational fluid dynamics (CFD) code, AMR-Wind, a structured-grid CFD code, and OpenFAST, a whole-turbine simulation code. In this poster we present the current status of the ExaWind software stack in the context of the modeling and simulation capabilities when the project started in 2016.

computational fluid dynamics↗
DOE OSTI2023

A single-tree algorithm to compute the Euclidean minimum spanning tree on GPUs

Computing the Euclidean minimum spanning tree (Emst) is a computationally demanding step of many algorithms. While work-efficient serial and multithreaded algorithms for computing Emst are known, designing an efficient GPU algorithm is challenging due to a complex branching structure, data dependencies, and load imbalances. In this paper, we propose a single-tree $Bor\mathring{u}vka$-based algorithm for computing Emst on GPUs. We use an efficient nearest neighbor algorithm and reduce the number of the required distance calculations by avoiding traversing subtrees with leaf nodes in the same component. The developed algorithms are implemented in a performance portable way using ArborX, an open-source geometric search library based on the Kokkos framework. We evaluate the proposed algorithm on various 2D and 3D datasets, show and compare it with the current state-of-the-art open-source CPU implementations. We demonstrate 4-24 × speedup over the fastest multi-threaded implementation. We prove the portability of our implementation by providing results on a variety of hardware: AMD EPYC 7763, Nvidia A100 and AMD MI250X. We show scalability of the implementation, computing Emst for 37 million 3D cosmological dataset in under a 0.5 second on a single A100 Nvidia GPU.

Prokopenko, Andrey↗
DOE OSTI2022

Five years of ForTrilinos ECP

The ForTrilinos subproject of the Exascale Computing Project (ECP) was initiated to bring the capabilities and scalability of the Trilinos numerical solver collection to Fortran scientific application codes. A novel Fortran extension to the Simplified Wrapper and Interface Generator (SWIG) tool, which automatically generates Fortran bindings from existing C/C++ library code, has been applied to key Trilinos solver libraries to generate the new ForTrilinos libraries. SWIG-Fortran has additionally been used to generate new Fortran compatibility layers for additional scientific libraries and applications. This report summarizes the products and impact of the ForTrilinos subproject.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

ExaWind: Exascale Predictive Wind Plant Flow Physics Modeling

The scientific goal of the ExaWind project is to advance our fundamental understanding of the flow physics governing whole wind plant performance, including wake formation, complex terrain impacts, and turbine-turbine-interaction effects. The primary application codes in the ExaWind environment are Nalu-Wind, an unstructured-grid computational fluid dynamics (CFD) code, AMR-Wind, a structured-grid CFD code, and OpenFAST, a whole-turbine simulation code. In this poster we present our progress towards simulating the ExaWind challenge problem, which is a predictive simulation of a wind farm with tens of megawatt-scale wind turbines dispersed over an area of 50 square kilometers.

49 EE - Wind and Water Power Program - Wind (EE-4W↗
DOE OSTI2020

ArborX: A Performance Portable Geometric Search Library

Searching for geometric objects that are close in space is a fundamental component of many applications. The performance of search algorithms comes to the forefront as the size of a problem increases both in terms of total object count as well as in the total number of search queries performed. Scientific applications requiring modern leadership-class supercomputers also pose an additional requirement of performance portability, i.e., being able to efficiently utilize a variety of hardware architectures. In this article, we introduce a new open-source C++ search library, ArborX, which we have designed for modern supercomputing architectures. Herein, we examine scalable search algorithms with a focus on performance, including a highly efficient parallel bounding volume hierarchy implementation, and propose a flexible interface making it easy to integrate with existing applications. We demonstrate the performance portability of ArborX on multi-core CPUs and GPUs and compare it to the state-of-the-art libraries such as Boost.Geometry.Index and nanoflann.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Automated Fortran–C++ Bindings for Large-Scale Scientific Applications

Although many active scientific codes use modern Fortran, most contemporary scientific software libraries are implemented in C and C++. Providing their numerical, algorithmic, or data management features to Fortran codes requires writing and maintaining substantial amounts of glue code. This paper introduces a tool that automatically generates native Fortran 2003 interfaces to C and C++ libraries. The tool supports C++ features that have no direct Fortran analog, such as templated functions and exceptions. Finally, a set of simple examples demonstrates the utility and scope of the tool, and timing measurements with a mock numerical library illustrate the minimal performance impact of the generated wrapper code.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Documenting automated Fortran-C++ bindings with SWIG

A new SWIG/Fortran tool introduced in Johnson et al. [to appear] automatically generates native Fortran2003 interfaces to C and C++ libraries. This allows a seamless integration of existing C++ libraries intothe Fortran applications. However, using the generated interfaces is complicated by the lack of appropriatedocumentation. In this report, we document a way to automatically port the existing Doxygendocumentation of the C++ libraries to Fortran. We use ForTrilinos library as our target application, anddiscuss the scope and limitations of this approach.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

MueLu User's Guide

This is the official user guide for MUELU multigrid library in Trilinos version 12.13 (Dev). This guide provides an overview of MUELU, its capabilities, and instructions for new users who want to start using MUELU with a minimum of effort. Detailed information is given on how to drive MUELU through its XML interface. Links to more advanced use cases are given. This guide gives information on how to achieve good parallel performance, as well as how to introduce new algorithms Finally, readers will find a comprehensive listing of available MUELU options. Any options not documented in this manual should be considered strictly experimental.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Deploy threading in Nalu solver stack

The goal of the ExaWind project is to enable predictive simulations of wind farms composed of many MW-scale turbines situated in complex terrain. Predictive simulations will require computational fluid dynamics (CFD) simulations for which the mesh resolves the geometry of the turbines, and captures the rotation and large deflections of blades. Whereas such simulations for a single turbine are arguably petascale class, multi-turbine wind farm simulations will require exascale-class resources. The primary code in the ExaWind project is Nalu, which is an unstructured-grid solver for the acousticallyincompressible Navier-Stokes equations, and mass continuity is maintained through pressure projection. The model consists of the mass-continuity Poisson-type equation for pressure and a momentum equation for the velocity. For such modeling approaches, simulation times are dominated by linear-system setup and solution for the continuity and momentum systems. For the ExaWind challenge problem, the moving meshes greatly affect overall solver costs as re-initialization of matrices and re-computation of preconditioners is required at every time step In this Milestone, we examine the effect of threading on the solver stack performance against flat-MPI results obtained from previous milestones using Haswell performance data full-turbine simulations. Whereas the momentum equations are solved only with the Trilinos solvers, we investigate two algebraic-multigrid preconditioners for the continuity equations: Trilinos/Muelu and HYPRE/BoomerAMG. These two packages embody smoothed-aggregation and classical Ruge-Stiiben AMG methods, respectively. In our FY18 Q2 report, we described our efforts to improve setup and solve of the continuity equations under flat-MPI parallelism. While significant improvement was demonstrated in the solve phase, setup times remained larger than expected. Starting with the optimized settings described in the Q2 report, we explore here simulation performance where OpenMP threading is employed in the solver stack. For Trilinos, threading is acheived through the Kokkos abstraction where, whereas HYPRE/BoomerAMG employs straight OpenMP. We examined results for our mid-resolution baseline turbine simulation configuration (229M DOF). Simulations on 2048 Haswell cores explored the effect of decreasing the number of MPI ranks while increasing the number of threads. Both HYPRE and Trilinos exhibited similar overal solution times, and both showed dramatic increases in simulation time in the shift from MPI ranks to OpenMP threads. This increase is attributed to the large amount of work per MPI rank starting at the single-thread configuration. Decreasing MPI ranks, while increasing threads, may be increasing simulation time due to thread synchronization and start-up overhead contributing to the latency and serial time in the model. These result showed that an MPI+OpenMP parallel decomposition will be more effective as the amount per MPI rank computation per MPI rank decreases and the communication latency increases. This idea was demonstrated in a strong scaling study of our low-resolution baseline model (29M DOF) with the Trilinos-HYPRE configuration. While MPI-only results showed scaling improvement out to about 1536 cores, engaging threading carried scaling improvements out to 4128 cores — roughly 7000 DOF per core. This is an important result as improved strong scaling is needed for simulations to be executed over sufficiently long simulated durations (i.e., for many timesteps). In addition to threading work described above, the team examined solver-performance improvements by exploring communication-overhead in the HYPRE-GMRES implementation through a communicationoptimal- GMRE algorithm (CO-GMRES), and offloading compute-intensive solver actions to GPUs. To those ends, a HYPRE mini-app was allow us to easily test different solver approaches and HYPRE parameter settings without running the entire Nalu code. With GPU acceleration on the Summitdev supercomputer, a 20x speedup was achieved for the overall preconditioner and solver execution time for the mini-app. A study on Haswell processors showed that CO-GMRES provides benefits as one increases MPI ranks.

17 WIND ENERGY↗