Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “High Performance Computing (HPC)”

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 1,117 records · Page 62

Scalable Energy Efficiency with Resilience for High Performance Computing Systems: A Quantitative Methodology

Energy efficiency and resilience are two crucial challenges for HPC systems to reach exascale. While energy efficiency and resilience issues have been extensively studied individually, little has been done to understand the interplay between energy efficiency and resilience for HPC systems. Decreasing the supply voltage associated with a given operating frequency for processors and other CMOS-based components can significantly reduce power consumption. However, this often raises system failure rates and consequently increases application execution time. In this work, we present an energy saving undervolting approach that leverages the mainstream resilience techniques to tolerate the increased failures caused by undervolting.

energy, resilience↗

Experiences with High-Level Programming Directives for Porting Applications to GPUs

HPC systems now exploit GPUs within their compute nodes to accelerate program performance. As a result, high-end application development has become extremely complex at the node level. In addition to restructuring the node code to exploit the cores and specialized devices, the programmer may need to choose a programming model such as OpenMP or CPU threads in conjunction with an accelerator programming model to share and manage the difference node resources. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. In order to offset the high development cost of creating CUDA or OpenCL kernels, directives have been proposed for programming accelerator devices, but their implications are not well known. In this paper, we evaluate the state of the art accelerator directives to program several applications kernels, explore transformations to achieve good performance, and examine the expressiveness and performance penalty of using high-level directives versus CUDA. We also compare our results to OpenMP implementations to understand the benefits of running the kernels in the accelerator versus CPU cores.

43 PARTICLE ACCELERATORS↗

A Conceptual Architecture for a Quantum-HPC Middleware

Abstract-Quantum computing promises potential for science and industry by solving certain computationally complex problems faster than classical computers. Quantum computing systems evolved from coupled to classical computing nodes (HPC). With the increasing monolithic systems towards modular architectures comprising multiple quantum processing units (QPUs) scale, middleware systems that facilitate the efficient coupling of quantum-classical computing are becoming critical. Through an in-depth analysis of quantum applications, integration patterns and systems, we identified a gap in understanding Quantum HPC middleware systems. We present a conceptual middleware to facilitate reasoning about quantum-classical integration and serve as the basis for a future middleware system. An essential contribution of this paper lies in leveraging well-established high performance computing abstractions for managing workloads, tasks, and resources to integrate quantum computing into HPC systems seamlessly

97 MATHEMATICS AND COMPUTING↗

2021 Supercomputer Institute Outbrief [Slides]

An ideal early career candidate for the HPC division is one who has a bachelor’s degree in a related field and experience with concepts of high performance computing. Most programs do not provide a good background on the systems side of HPC in their undergraduate curricula. We are competing for the same candidates that large tech companies are interested in and recruiting into a field that isn’t on the radar of many undergrads.

97 MATHEMATICS AND COMPUTING↗

High Performance Computing Modeling Advances Accelerator Science for High-Energy Physics

The development and optimization of particle accelerators are essential for advancing our understanding of the properties of matter, energy, space, and time. Particle accelerators are complex devices whose behavior involves many physical effects on multiple scales. Therefore, advanced computational tools utilizing high-performance computing are essential for accurately modeling them. In the past decade, the US Department of Energy's SciDAC program has produced accelerator-modeling tools that have been employed to tackle some of the most difficult accelerator science problems. The authors discuss the Synergia framework and its applications to high-intensity particle accelerator physics. Synergia is an accelerator simulation package capable of handling the entire spectrum of beam dynamics simulations. Our authors present Synergia's design principles and its performance on HPC platforms.

43 PARTICLE ACCELERATORS↗

Overview of HPC at LANL

This report provides an overview of the High Performance Computing Division and activities at LANL.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Application Experiences on a GPU-Accelerated Arm-based HPC Testbed

This paper assesses and reports the experience of ten teams working to port, validate, and benchmark several High Performance Computing applications on a novel GPU-accelerated Arm testbed system. The testbed consists of eight NVIDIA Arm HPC Developer Kit systems, each one equipped with a server-class Arm CPU from Ampere Computing and two data center GPUs from NVIDIA Corp. The systems are connected together using InfiniBand interconnect. The selected applications and mini-apps are written using several programming languages and use multiple accelerator-based programming models for GPUs such as CUDA, OpenACC, and OpenMP offloading. Working on application porting requires a robust and easy-to-access programming environment, including a variety of compilers and optimized scientific libraries. The goal of this work is to evaluate platform readiness and assess the effort required from developers to deploy well-established scientific workloads on current and future generation Arm-based GPU-accelerated HPC systems. The reported case studies demonstrate that the current level of maturity and diversity of software and tools is already adequate for large-scale production deployments.

Elwasif, Wael↗

High Performance GPU-Based FEM Code for Simulating Welding and Additive Manufacturing

An in-house HPC finite element code with novel computational schemes provides fast and high-fidelity simulation capability for welding processes and allied manufacturing processes such as metal additive manufacturing. Benchmark comparison confirmed that, for the testing cases studied, the newly developed in-house code resulted in increases in computational speed in the ranges of 100x to 1000x depending on the size and complexity of the testing cases, when compared to today's commercial welding process simulation tools. For a reference problem in nuclear reactor repair welding, ORNL's in-house high fidelity code was able to complete the simulation in 3.2 days, which would take estimated 10 to 20 years using today's commercial code. Similar order of magnitudes of reduction in computational times for metal additive manufacturing process were also achieved.

Feng, Zhili↗

High-Performance Computing for Real-Time Grid Analysis and Operation

Power grids worldwide are undergoing an unprecedented transition as a result of grid evolution meeting information revolution. The grid evolution is largely driven by the desire for green energy. Emerging grid technologies such as renewable generation, smart loads, plug-in hybrid vehicles, and distributed generation provide opportunities to generate energy from green sources and to manage energy use for better system efficiency. With utility companies actively deploying these technologies, a high level of penetration of these new technologies is expected in the next 5-10 years, bringing in a level of intermittency, uncertainties, and complexity that the grid did not see nor design for. On the other hand, the information infrastructure in the power grid is being revolutionized with large-scale deployment of sensors and meters in both the transmission and distribution networks. The future grid will have two-way flows of both electrons and information. The challenge is how to take advantage of the information revolution: pull the large amount of data in, process it in real time, and put information out to manage grid evolution. Without addressing this challenge, the opportunities in grid evolution will remain unfulfilled. This transition poses grand challenges in grid modeling, simulation, and information presentation. The computational complexity of underlying power grid modeling and simulation will significantly increase in the next decade due to an increased model size and a decreased time window allowed to compute model solutions. High-performance computing is essential to enable this transition. The essential technical barrier is to vastly increase the computational speed so operation response time can be reduced from minutes to seconds and sub-seconds. The speed at which key functions such as state estimation and contingency analysis are conducted (typically every 3-5 minutes) needs to be dramatically increased so that the analysis of contingencies is both comprehensive and real time. An even bigger challenge is how to incorporate dynamic information into real-time grid operation. Today’s online grid operation is based on a static grid model and can only provide a static snapshot of current system operation status, while dynamic analysis is conducted offline because of low computational efficiency. The offline analysis uses a worst-case scenario to determine transmission limits, resulting in under-utilization of grid assets. This conservative approach does not necessarily lead to reliability. Many times, actual power grid scenarios are not studied, and they will push the grid over the edge and resulting in outages and blackouts. This chapter addresses the HPC needs in power grid analysis and operations. Example applications such as state estimation and contingency analysis are given to demonstrate the value of HPC in power grid applications. Future research directions are suggested for high performance computing applications in power grids to improve the transparency, efficiency, and reliability of power grids.

Contingency Analysis↗

Experiences with High-Level Programming Directives for Porting Applications to GPUs. In: Facing the Multicore--Challenge II, Lecture Notes in Computer Science.

HPC systems now exploit GPUs within their compute nodes to accelerate program performance. As a result, high-end application development has become extremely complex at the node level. In addition to restructuring the node code to exploit the cores and specialized devices, the programmer may need to choose a programming model such as OpenMP or CPU threads in conjunction with an accelerator programming model to share and manage the different node resources. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. In order to offset the high development cost of creating CUDA or OpenCL kernels, directives have been proposed for programming accelerator devices, but their implications are not well known. In this paper, we evaluate the state of the art accelerator directives to program several applications kernels, explore transformations to achieve good performance, and examine the expressivity and performance penalty of using high-level directives versus CUDA. We also compare our results to OpenMP implementations to understand the benefits of running the kernels in the accelerator versus CPU cores.

CUDA Implementation↗

Exploiting HPC Platforms for Metagenomics: Challenges and Opportunities (MICW - Metagenomics Informatics Challenges Workshop: 10K Genomes at a Time)

DOE JGI's Zhong Wang, chair of the High-performance Computing session, gives a brief introduction before Berkeley Lab's Shane Canon talks about "Exploiting HPC Platforms for Metagenomics: Challenges and Opportunities" at the Metagenomics Informatics Challenges Workshop held at the DOE JGI on October 12-13, 2011.

59 BASIC BIOLOGICAL SCIENCES↗

Understanding HPC Benchmark Performance on Intel Broadwell and Cascade Lake Processors

Hardware platforms in high performance computing are constantly getting more complex to handle even when considering multicore CPUs alone. Numerous features and configuration options in the hardware and the software environment that are relevant for performance are not even known to most application users or developers. Microbenchmarks, i.e., simple codes that fathom a particular aspect of the hardware, can help to shed light on such issues, but only if they are well understood and if the results can be reconciled with known facts or performance models. The insight gained from microbenchmarks may then be applied to real applications for performance analysis or optimization. In this paper we investigate two modern Intel x86 server CPU architectures in depth: Broadwell EP and Cascade Lake SP. We highlight relevant hardware configuration settings that can have a decisive impact on code performance and show how to properly measure on-chip and off-chip data transfer bandwidths. The new victim L3 cache of Cascade Lake and its advanced replacement policy receive due attention. Finally we use DGEMM, sparse matrix-vector multiplication, and the HPCG benchmark to make a connection to relevant application scenarios.

97 MATHEMATICS AND COMPUTING↗

The Trinity Center of Excellence Co-Design Best Practices

Co-design between application developers and the vendors providing high-performance computing systems is essential to efficiently using current-generation HPC systems and shaping next-generation architectures. Application developers are faced with the challenge of effectively orchestrating many hardware threads and exploiting vector instructions, which can require code modifications. Will refactoring applications for current-generation systems apply to future-generation systems? Is there help available for this enormous task? Over the past 10 years, a partnership has evolved to address these issues. Centers of Excellence are mutually beneficial collaborations between application developers and vendors; vendors receive a better understanding of the needs of their users, and users receive a better understanding of the characteristics of the hardware, system software, and tools available to address their challenges. Lastly, this article describes the partnership between two vendors, Cray and Intel, and the application teams working together in the Trinity Center of Excellence.

97 MATHEMATICS AND COMPUTING↗

Failure prediction using machine learning in a virtualised HPC system and application

Failure is an increasingly important issue in high performance computing and cloud systems. As large-scale systems continue to grow in scale and complexity, mitigating the impact of failure and providing accurate predictions with sufficient lead time remains a challenging research problem. Traditional existing fault-tolerance strategies such as regular check-pointing and replication are not adequate because of the emerging complexities of high performance computing systems. This necessitates the importance of having an effective as well as proactive failure management approach in place aimed at minimizing the effect of failure within the system. With the advent of machine learning techniques, the ability to learn from past information to predict future pattern of behaviours makes it possible to predict potential system failure more accurately. Thus, in this paper, we explore the predictive abilities of machine learning by applying a number of algorithms to improve the accuracy of failure prediction. We have developed a failure prediction model using time series and machine learning, and performed comparison based tests on the prediction accuracy. The primary algorithms we considered are the support vector machine (SVM), random forest (RF), k-nearest neighbors (KNN), classification and regression trees (CART) and linear discriminant analysis (LDA). Experimental results indicates that the average prediction accuracy of our model using SVM when predicting failure is 90% accurate and effective compared to other algorithms. This finding implies that our method can effectively predict all possible future system and application failures within the system.

Mohammed, Bashir↗

Exploiting communication concurrency on high performance computing systems

Although logically available, applications may not exploit enough instantaneous communication concurrency to maximize hardware utilization on HPC systems. This is exacerbated in hybrid programming models such as SPMD+OpenMP. We present the design of a "multi-threaded" runtime able to transparently increase the instantaneous network concurrency and to provide near saturation bandwidth, independent of the application configuration and dynamic behavior. The runtime forwards communication requests from application level tasks to multiple communication servers. Our techniques alleviate the need for spatial and temporal application level message concurrency optimizations. Experimental results show improved message throughput and bandwidth by as much as 150% for 4KB bytes messages on InfiniBand and by as much as 120% for 4KB byte messages on Cray Aries. For more complex operations such as all-to-all collectives, we observe as much as 30% speedup. This translates into 23% speedup on 12,288 cores for a NAS FT implemented using FFTW. We also observe as much as 76% speedup on 1,500 cores for an already optimized UPC+OpenMP geometric multigrid application using hybrid parallelism.

97 MATHEMATICS AND COMPUTING↗

A high resolution agent-based model to support walk-bicycle infrastructure investment decisions: A case study with New York City

Active transportation modes--walk and bicycle--are central for low carbon transport, healthy living, and complete streets initiative. Building a community with amenable walk and bicycle facilities asks for smart planning and investments. It is critical to investigate the impact of infrastructure building or expansion on the overall walk and bicycle mode usage prior to making investment choices utilizing public tax money. This research developed an agent-based model to support investment decisions that allows to assess the impact of changes in walk-bike infrastructures at a high spatial resolution (e.g., block group level). The agent-based model (ABM) utilizes data from a synthetic population simulator generating agents with corresponding socio-demographic characteristics, and integrates facility attributes regarding walking and bicycling (e.g., sidewalk width, bike lane length) into the mode choice decision making process. Moreover, the ABM accounts for the effect of social interactions among agents who live and work at the same geographic locations. Finally, GIS-based maps are developed at block group resolution that allows exploring the effect of walk-bike infrastructure related investments. The results from New York City case study indicate that infrastructure investments such as widening sidewalk and increasing bike lane network can positively influence the active transportation mode choices. In addition, the level of impact varies with geographic locations--different boroughs of New York City will have different impacts. Lastly, social promotions resulting in higher social interaction among agents can reinforce the impacts of infrastructure changes.

97 MATHEMATICS AND COMPUTING↗

High Performance Computing tutorials

This source code includes tutorials targeting a general audience that will help users to use high performance computing environments and systems at Livermore Computing and beyond. Examples of HPC tools covered include mpi and posix. Tutorials may include source code demonstrating usage.

Herriman, JaneE.↗