Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “HPC performance modeling”

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 469 records · Page 26

HIPLZ: Enabling performance portability for exascale systems

While heterogeneous computing has emerged as a dominant trend in current and future High-Performance Computing (HPC) systems, it is also widely recognized that this shift has led to increased software complexity due to a proliferation of programming systems for different heterogeneous processors. One such example is the Heterogeneous-Compute Interface for Portability from AMD (HIP ), which is composed of a C Runtime API and C++ Kernel Language. Many HPC applications will likely use HIP on future exascale systems (e.g., Frontier and El Capitan), but HIP currently only targets AMD and NVIDIA processors. This limitation creates challenges for users who would also like to run their applications on exascale systems based on other architectures (e.g., Aurora, which is based on Intel hardware) that are currently not targeted by HIP . In this paper, we introduce the design and implementation of HIPLZ , a compiler and runtime system that uses the Intel Level Zero API to support HIP on Intel GPU architectures. We discuss the design of HIPLZ , derived from HIPCL (an implementation of HIP on top of OpenCL ), and portability issues that occur from using the Level Zero runtime as a backend. We evaluate our implementation by running several performance benchmarks and mini-apps written in HIP on Intel architectures using HIPLZ . Our results show that this approach provides competitive performance relative to Intel's OpenCL implementations on Intel Gen9 and UHD Graphics 770 GPUs, while providing good coverage of features needed by HPC applications. Overall, this approach is a promising demonstration of enabling performance portability for exascale systems.

97 MATHEMATICS AND COMPUTING↗

In Situ Visualization of Radiation Transport Geometry

The ultimate goal for radiation transport is to perform full-core reactor modelling and simulation. Advances in computational simulation bring this goal close to reality and the newest Monte Carlo transport codes have begun to shift to using accelerators that have become a stalwart in the supercomputing and HPC space. Within the reactor modelling and simulation community, Monte Carlo transport simulations are considered the gold standard for simulation. Through the use of "combinatorial geometry" (constructive solid geomtry), complex models can be used with fewer approximation compromises while at the same time scale to run on some of the largest supercomputers in the world.Unfortunately, the state-of-the-art for "combinatorial geometry" visualization is to decompose the geometry into a mesh. This approach could require a significant amount of memory which is antithetical to in situ visualization. To address this issue, we introduce a ray caster for visualizing combinatorial geometry in radiation transport code. By only using the accelerators for the radiation transport code and leaving the CPU cores idle, there is an opportunity to conduct on node in situ visualization with the idle CPU cores, something domain experts have up to this point been unable to do. By utilizing VTK-m, the visualization can be run on the CPU as this particular application demands, but also run on any architecture that is supported by VTK-m, enabling future re-use across different platforms.

Kim, Mark B.↗

Hybrid PDES Simulation of HPC Networks Using Zombie Packets

Although high-fidelity network simulations have proven to be reliable and cost-effective tools to peer into architectural questions for high-performance computing (HPC) networks, they incur a high resource cost. The time spent in simulating a single millisecond of network traffic in the highest detail can take hours, even for static, well-behaved traffic patterns such as uniform random. Surrogate models offer a significant reduction in runtime, yet they cannot serve as complete replacements and should only be used when appropriate. Thus, there is a need for hybrid modeling, where high-fidelity simulation and surrogates run side-by-side. Here, we present a surrogate model for HPC networks in which: packets bypass the network, while the network state is left untouched, i.e., suspended. To bypass the network, we use historical data to estimate the arrival time at which every packet should be scheduled at; to suspend the network, all in-flight packets are scheduled to arrive at their destinations, and are kept in the system to awaken as zombies when switching back to high-fidelity. Speedup for a hybrid model is relative to the proportion of surrogate to high-fidelity. This light-weight surrogate obtained up to 76× speedup. Keeping the zombies in the network showed an increase in the accuracy of the high-fidelity simulation on restart when compared to restarting the network from an empty state.

HPC networks↗

Towards Low-Overhead Resilience for Data Parallel Deep Learning

Data parallel techniques have been widely adopted both in academia and industry as a tool to enable scalable training of deep learning models. At scale, DL training jobs can fail due to software or hardware bugs, may need to be preempted or terminated due to unexpected events, or may perform suboptimally because they were misconfigured. Under such circumstances, there is a need to recover and/or reconfigure data-parallel DL training jobs on-the-fly, while minimizing the impact on the accuracy of the DNN model and the runtime overhead. In this regard, state-of-art techniques adopted by the HPC community mostly rely on checkpoint-restart, which inevitably leads to loss of progress, thus increasing the runtime overhead. In this paper we explore alternative techniques that exploit the properties of modern deep learning frameworks (overlapping of gradient averaging and weight updates with local gradient computations through pipeline parallelism) to reduce the overhead of resilience/elasticity. To this end we introduce a failure simulation framework and two resilience strategies (immediate mini-batch rollback and lossy forward recovery), which we study compared with checkpoint-restart approaches in a variety of settings in order to understand the trade-offs between the accuracy loss of the DNN model and the runtime overhead.

data-parallel training↗

Matrix Factorizations at Scale: a Comparison of Scientific Data Analytics in Spark and C+MPI Using Three Case Studies

We explore the trade-offs of performing linear algebra using Apache Spark, compared to traditional C and MPI implementations on HPC platforms. Spark is designed for data analytics on cluster computing platforms with access to local disks and is optimized for data-parallel tasks. We examine three widely-used and important matrix factorizations: NMF (for physical plausibility), PCA (for its ubiquity) and CX (for data interpretability). We apply these methods to 1.6TB particle physics, 2.2TB and 16TB climate modeling and 1.1TB bioimaging data. The data matrices are tall-and-skinny which enable the algorithms to map conveniently into Spark’s data parallel model. We perform scaling experiments on up to 1600 Cray XC40 nodes, describe the sources of slowdowns, and provide tuning guidance to obtain high performance.

Gittens, Alex↗

High Performance Computing Data Center [Fact Sheet]

The high performance computing (HPC) data center in the Energy Systems Integration Facility (ESIF) is designed to be one of the most energy efficient data centers in the world, featuring warm-water liquid cooling and waste heat capture and re-use. Its HPC capabilities support the breadth of research at NREL, leading to increased efficiency and lower costs for important technologies including solar photovoltaics, wind energy, energy storage, electric vehicles, and the large-scale integration of renewables with the grid. With its state-of-the-art computational modeling and predictive simulation capabilities, the HPC data center will reduce the risks and uncertainty that are often barriers to industry adopting new and innovative technologies, thereby accelerating the transformation of our nation’s energy system.

97 MATHEMATICS AND COMPUTING↗

Scalable Data-Intensive Geocomputation: A Design for Real-Time Continental Flood Inundation Mapping

The convergence of data-intensive and extreme-scale computing enables an integrated software and data ecosystem for scientific discovery. Developments in this realm will fuel transformative research in data-driven interdisciplinary domains. Geocomputation provides computing paradigms in Geographic Information Systems (GIS) for interactive computing of geographic data, processes, models, and maps. Because GIS is data-driven, the computational scalability of a geocomputation workflow is directly related to the scale of the GIS data layers, their resolution and extent, as well as the velocity of the geo-located data streams to be processed. Unique in high user interactivity and low end-to-end latency requirements, geocomputation applications will dramatically benefit from the convergence of high-end data analytics (HDA) and high-performance computing (HPC). The application level challenge, however, is to identify and eliminate computational bottlenecks that arise along a geocomputation workflow. Indeed, poor scalability at any of the workflow components is detrimental to the entire end-to-end pipeline. Here, we study a large geocomputation use case in flood inundation mapping that handles multiple national-scale geospatial datasets and targets low end-to-end latency. We discuss benefits and challenges for harnessing both HDA and HPC for data-intensive geospatial data processing and intensive numerical modeling of geographic processes. We propose an HDA+HPC geocomputation architecture design that couples HDA (e.g., Spark)-based spatial data handling and HPC-based parallel data modeling. Key techniques for coupling HDA and HPC to bridge the two different software stacks are reviewed and discussed.

Liu, Yan↗

A Quantitative Study of Deep Learning Training on Heterogeneous Supercomputers

Deep learning (DL) has become a key technique for solving complex problems in scientific research and discovery. DL training for science is substantially challenging because it has to deal with massive quantities of multi-dimensional data. High- performance computing (HPC) supercomputers are increasingly being employed for meeting the exponentially growing demand for DL. Multiple GPUs and high-speed interconnect network are needed for supporting DL on HPC systems. However, the excessive use of GPUs without considering effective benefits leads to inefficient resource utilization of these expensive setups. In this paper, we conduct a quantitative analysis to gauge the efficacy of DL workloads on the latest HPC system and identify viability of next-generation DL-optimized heterogeneous supercomputers for enabling researchers to develop more efficient resource management and distributed DL middleware. We evaluate well- known DL models with large-scale datasets using the popular TensorFlow framework, and provide a thorough evaluation in- cluding scalability, accuracy, variability, storage resource, GPU- GPU/GPU-CPU data transfer, and GPU utilization. Our analysis reveals that the latest heterogeneous supercomputing cluster shows varying performance trend as compared to the existing literature for single- and multi-node training. To the best of our knowledge, this is the first work to conduct such a quantitative and comprehensive study of DL training on a supercomputing system with multiple GPUs.

Han, Jingoo↗

Scaling Ensembles of Data-Intensive Quantum Chemical Calculations for Millions of Molecules

Deep learning models are efficient computational tools that can accelerate the inverse design of molecules with desired functional properties by generating predictions at a fraction of the time required by traditional quantum chemical approaches. To ensure that a model maintains accuracy and transferability across broad regions of the chemical space explored during the inverse design, it must be trained on massively large volumes of simulation data. This requires running large-scale ensemble quantum chemical calculations on high-performance computing (HPC) systems for data collection. However, the efficient execution of such large ensemble calculations and the management of large volumes of output data require tools that can judiciously utilize computational resources and manage metadata overhead on the file system. Therefore, we present a high-performance, scalable, ensemble management framework for performing data-intensive quantum chemical electronic structure calculations for organic molecules. This framework provides abstractions to plug different ab initio, first principles, and first principles-based semi-empirical methods and executes them efficiently at large scale on HPC systems. It dynamically distributes tasks to resources and uses tiered storage for managing large collections of files. We employed this framework to process over ten million organic molecules and generate open-source datasets that provide UV-vis absorption spectra by running time-dependent density-functional tight-binding calculations. It is the largest database containing molecular optical spectra that were simulated with quantum chemical methods in a consistent manner.

Mehta, Kshitij↗

Evaluating Burst Buffer Placement in HPC Systems

Burst buffers (BBs) are increasingly exploited in contemporary supercomputers to bridge the performance gap between compute and storage systems. The design of BBs, particularly the placement of these devices and the underlying network topology, impacts both performance and cost. As the cost of other components such as memory and accelerators is increasing, it is becoming more important that HPC centers provision BBs tailored to their workloads.This work contributes a provisioning system to provide accurate, multi-tenant simulations that model realistic application and storage workloads from HPC systems. The framework aids HPC centers in modeling their workloads against multiple network and BB configurations rapidly. In experiments with our framework, we provide a comparison of representative Oak Ridge Leadership Computing Facility (OLCF) I/O workloads against multiple BB designs. We analyze the impact of these designs on latency, I/O phase lengths, contention for network and storage devices, and choice of network topology.

Khetawat, Harsh↗

PV Hosting Capacity Estimation: Experiences with Scalable Framework; Preprint

Hosting capacity is an indication of the amount of photovoltaics (PV) can be hosted in a distribution system. This paper presents a framework for estimating distributed PV hosting capacity at scale. We first analyze the key challenges of performing relevant large scale simulation including computational and modeling challenges. Then, we develop two python-based software tools in order to conduct repeatable distribution analyses: Distribution Integration Solution Cost Options (DISCO) for configuring and analyzing simulations, and JADE for parallelizing jobs on HPC clusters. A case study of hosting capacity estimation for SMART-DS SFO 2000+ synthetic feeders is used to demonstrate the capability of the developed framework and tools. The framework and tools can help utilities assess the overall hosting capacity of their service territory, through which the overall upgrade cost can be better planned in order to integrate more PV in the future.

distributed energy resources↗

Static Graphs for Coding Productivity in OpenACC

The main contribution of this work is to increase the coding productivity for GPU programming by using the concept of Static Graphs. To do so, we have combined the new CUDA Graph API with the OpenACC programming model. We use as test cases a well-known and widely used problems in HPC and AI: the Particle Swarm Optimization. We complement the OpenACC functionality with the use of CUDA Graph, achieving accelerations of more than one order of magnitude, and a performance very close to a reference and optimized CUDA code. Finally, we propose a new specification to incorporate the concept of Static Graphs into the OpenACC specification.

Toledo, Leonel↗

Privacy Preservation from High-Performance Computing to Autonomous Science [Industrial and Governmental Activities]

High-Performance Computing (HPC) and Leadership-Class Supercomputing are driving forces behind scientific advancements, enabling researchers to tackle complex challenges in physics, chemistry, biology, and engineering. These systems power vast simulations and data analyses, fueling discoveries in fields ranging from materials science to climate modeling. However, their use often involves processing sensitive data—such as proprietary industry simulations, biomedical records, and national security computations—posing significant privacy concerns. In conclusion, this issue is amplified in collaborative environments like Department of Energy (DOE) user facilities, where HPC resources are shared across institutions to foster innovation.

Kotevska, Olivera [Oak Ridge National Laboratory (↗

Connecting Performance Analysis and Visualization to Advance Extreme Scale Computing

In the first week of January 2014 Dagstuhl hosted a Perspectives Workshop on “Connecting Performance Analysis and Visualization to Advance Extreme Scale Computing”. The event brought together two previously separate communities – from Visualization and HPC Performance Analysis – to discuss a long term joined research agenda. The goal was to identify and address the challenges in using visual representations to understand and optimize the performance of extreme-scale applications running on today’s most powerful computing systems like climate modeling, combustion, material science or astro-physics simulations.

97 MATHEMATICS AND COMPUTING↗

Optimization of Scrap Melting Using an Electric Arc in Steel Manufacturing

Steel industry is crucial to the national economy and security. Around 67% of crude steel in the U.S is produced in electric arc furnaces (EAF), which is energy intensive. Around 140 EAFs operate in the U.S., consuming about 8.6x10 7 MMBtu/year of electricity. One of major challenges for EAFs includes maximizing the efficiency of the electrical energy provided in the form of electric arcs to melt various scrap mixes. To address this issue, a computational fluid dynamics (CFD) methodology is chosen to analyze scrap melting using the electric arc. Due to complex furnace phenomena and the wide variety of potential scenarios, high performance computing (HPC) is essential to yield comprehensive and detailed CFD analyses and systematic parametric studies for optimized EAF operation. The objectives are to 1) simulate scrap melting using electric arc, 2) evaluate electrode/arc position for optimum scrap melting and 3) establish reduced order model for CFD data-base for fast model calculation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

The Virtual Blast Furnace - An Integrated High Performance Computing Modeling, Simulation, and Visualization Capability for Steel Manufacturing (Final Report)

Many manufacturing industries require substantial capital and utilize energy intensive processes that involve complex phenomena. One example of such an industry is the steel industry, which is the fourth largest energy consuming industry in the U.S. By harnessing the power of High-Performance Computing (HPC) to enhance current simulation and visualization methods in the steel industry, it should be possible to increase resolution and/or decrease time of these methods by a factor of 1000. In this way, information can be obtained in a time frame that is useful for making business and engineering decisions, optimizing manufacturing processes and, ultimately, improving the completeness of U.S. industries. For example, if coke usage in blast furnaces were optimized such that the average coke rate was reduced from 797 lb/net tonne of hot metal (NTHM) to 604 lb/NTHM, costs could be reduced by $894 million/year. Additionally, members of the steel industry need the flexibility to efficiently operate blast furnaces at a range of production rates in order to meet fluctuating market demands. Large scale parameter studies can be utilized to discover workable operating parameters at a range of production rates.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Implementation of Feldman-Cousins Corrections and Oscillation Calculations in the HPC Environment for the NOvA Experiment

Analysis of neutrino oscillation data involves a combination of complex fitting procedures and statistical correction techniques that are used to determine the full three-flavor PMNS parameters and constraint contours. These techniques rely on computationally intensive “multi-universe” stochastic modeling. The process of calculating these contours and corrections can dominate final stages of the data analysis and become a bottleneck for examining the ef-fect of systematic variations on the final results. As part of the DOE SciDAC-4 sponsored research program, we present a new implementation of a neutrino oscillation fitting and framework to carry out calculations of Feldman-Cousins corrections. This implementation is specifically designed and optimized to operate on modern High-Performance Computing (HPC) facilities. We present the performance of the system in calculating allowed regions for the NOvA experiment based on 8.9 x 1020 and 6.9 x 1020 protons-on-target (POT) neutrino and antineutrino datasets, and compare the performance of this new implementation run at the NERSC supercomputing facility with that from methods used previously by the NOvA collaboration running on grid computing facilities.

Sousa, Alexandre↗

Enabling Fortran Standard Parallelism in GAMESS for Accelerated Quantum Chemistry Calculations

The performance of Fortran 2008 DO CONCURRENT (DC) relative to OpenACC and OpenMP target offloading (OTO) with different compilers is studied for the GAMESS quantum chemistry application. Specifically, DC and OTO are used to offload the Fock build, which is a computational bottleneck in most quantum chemistry codes, to GPUs. The DC Fock build performance is studied on NVIDIA A100 and V100 accelerators and compared with the OTO versions compiled by the NVIDIA HPC, IBM XL, and Cray Fortran compilers. The results show that DC can speed up the Fock build by 3.0× compared with that of the OTO model. Finally, with similar offloading efforts, DC is a compelling programming model for offloading Fortran applications to GPUs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗