High-Performance Molecular Dynamics Simulation for Biological and Materials Sciences: Challenges of Performance Portability

Sedova, Ada; Eblen, John D.; Budiardja, Reuben D.; Tharrington, Arnold; Smith, Jeremy C.

doi:10.1109/p3hpc.2018.00004

Cited by 20 publications

(18 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As discussed in sections 6.1.2 , 6.2.2 , and 6.2.4 , there are a number of open-source molecular simulation packages for which several benchmark and case studies have been produced. 195 , 215 , 263 However, this has not been done for process simulation packages mainly because the majority of these software are not available as open-source. In fact, among all the software introduced in section 6.3.6 , only one computer code has been released as open source.…”

Section: Challenges Of Multiscale Materials Screening: Accuracy Data Availability and Reproducibilitymentioning

confidence: 99%

Performance-Based Screening of Porous Materials for Carbon Capture

et al. 2021

View full text Add to dashboard Cite

Computational screening methods have changed the way new materials and processes are discovered and designed. For adsorption-based gas separations and carbon capture, recent efforts have been directed toward the development of multiscale and performance-based screening workflows where we can go from the atomistic structure of an adsorbent to its equilibrium and transport properties at different scales, and eventually to its separation performance at the process level. The objective of this work is to review the current status of this new approach, discuss its potential and impact on the field of materials screening, and highlight the challenges that limit its application. We compile and introduce all the elements required for the development, implementation, and operation of multiscale workflows, hence providing a useful practical guide and a comprehensive source of reference to the scientific communities who work in this area. Our review includes information about available materials databases, state-of-the-art molecular simulation and process modeling tools, and a complete catalogue of data and parameters that are required at each stage of the multiscale screening. We thoroughly discuss the challenges associated with data availability, consistency of the models, and reproducibility of the data and, finally, propose new directions for the future of the field.

show abstract

Section: Challenges Of Multiscale Materials Screening: Accuracy Data Availability and Reproducibilitymentioning

confidence: 99%

Performance-Based Screening of Porous Materials for Carbon Capture

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The miniapp was designed based on two existing versions of the full AutoDock-GPU program: the new CUDA version developed with the help of NVIDIA, 2 as well as a Kokkos version. 3 Both of these were created in an effort to provide a version of the original GPU-accelerated AutoDock-GPU program, developed by Scripps Research in OpenCL, 4 that could be deployed on Summit for drug discovery efforts against COVID-19. There are currently no NVIDIA drivers for OpenCL on Power9, which prevents the original version of the program from running on Summit.…”

Section: B Design and Structurementioning

confidence: 99%

“…Therefore, the CUDA version currently represents the best performing version of this code. The performance of all portable solutions can thus be compared to the performance of this version to check how much performance is lost by supporting multiple platforms [4]. In addition, worker time spent optimizing portable solutions to enable improved performance on each platform, while remaining generally usable, can be counted and included in measures of performance portability [2].…”

Section: B Design and Structurementioning

confidence: 99%

“…The primary GPU vendor used in HPC has been NVIDIA, and the CUDA application programming interface (API) has been one of the most popular programming interfaces for generation of highly optimized GPU kernels. With the recent new emergence of other GPU vendors competing for spots in leadership computing, as well as new developments in CPU architectures that are promising that purely CPU-based supercomputers will continue to be competitive in top HPC categories, the need for portable programming models is becoming increasingly urgent to maintain productivity for computational scientists [1]- [4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Performance Portability of Molecular Docking Miniapp On Leadership Computing Platforms

Thavappiragasam

Scheinberg²,

Elwasif

et al. 2020

2020 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)

Self Cite

View full text Add to dashboard Cite

Rapidly changing computer architectures, such as those found at high-performance computing (HPC) facilities, present the need for mini-applications (miniapps) that capture essential algorithms used in large applications to test program performance and portability, aiding transitions to new systems. The COVID-19 pandemic has fueled a flurry of activity in computational drug discovery, including the use of supercomputers and GPU acceleration for massive virtual screens for therapeutics. Recent work targeting COVID-19 at the Oak Ridge Leadership Computing Facility (OLCF) used the GPU-accelerated program AutoDock-GPU to screen billions of compounds on the Summit supercomputer. In this paper we present the development of a new miniapp, miniAutoDock-GPU, that can be used to evaluate the performance and portability of GPU-accelerated proteinligand docking programs on different computer architectures. These tests are especially relevant as facilities transition from petascale systems and prepare for upcoming exascale systems that will use a variety of GPU vendors. The key calculations, namely, the Lamarckian genetic algorithm combined with a local search using a Solis-Wets based random optimization algorithm, are implemented. We developed versions of the miniapp using several different programming models for GPU acceleration, including a version using the CUDA runtime API for NVIDIA GPUs, and the Kokkos middle-ware API which is facilitated by C++ template libraries. A third version, currently in progress, uses the HIP programming model. These efforts will help facilitate the transition to exascale systems for this important emerging HPC application, as well as its use on a wide range of heterogeneous platforms.Index Terms-heterogeneous system, high-performance computing, performance portability, hybrid parallel programming model, molecular docking, drug discovery.

show abstract

“…Sedova et al expand this metric to consider architecture specialisation of certain key routines of applications [7]. In this study, we aim to use the same code base on all platforms, but we do discuss two mini-apps which require some algorithmic differences between different classes of architecture.…”

Section: A Related Workmentioning

confidence: 99%

Performance Portability across Diverse Computer Architectures

Deakin

McIntosh–Smith

Price

et al. 2019

2019 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)

View full text Add to dashboard Cite

Previous studies into performance portability have typically analysed a single application (and its various implementations) in isolation. In this study we explore the wider landscape of performance portability by considering a number of applications from across the space of dwarfs, written in multiple parallel programming models, and across a diverse set of architectures. We apply rigorous performance portability metrics, as defined by Pennycook et al [1]. We believe this is the broadest and most rigorous performance portability study to date, representing a far reaching exploration of the state of performance portability that is achievable today. We will present a summary of the performance portability of each application and programming model across our diverge range of twelve computer architectures, including six different server CPUs from five different vendors, five different GPUs from two different vendors, and one vector architecture. We will conclude with an analysis of the performance portability of key programming models in general, across different application spaces as well across differing architectures, allowing us to comment on more general performance portability principles.

show abstract

High-Performance Molecular Dynamics Simulation for Biological and Materials Sciences: Challenges of Performance Portability

Cited by 20 publications

References 58 publications

Performance-Based Screening of Porous Materials for Carbon Capture

Performance-Based Screening of Porous Materials for Carbon Capture

Performance Portability of Molecular Docking Miniapp On Leadership Computing Platforms

Performance Portability across Diverse Computer Architectures

Contact Info

Product

Resources

About