Engineering topics

Yin, Junqi

Publications and source records attributed to Yin, Junqi.

At least 37 records · Page 2

DOE OSTI2023

Evaluating the Cloud for Capability Class Leadership Workloads

Cloud platforms offer a variety of benefits that are very appealing for a large scale HPC facility with a diverse and dynamic user base and workload set. At the same time, there is cause for concern about transitioning to the cloud. Incorporating cloud resources into existing HPC facilities or even fully transitioning to a cloud deployment poses significant challenges at the technical, organizational, and economic levels. Regardless, based on current trends it is highly likely that cloud platforms will become an integral component of many HPC centers in some form. To gain a better understanding of both the limitations and capabilities of current cloud infrastructures we evaluated the public offerings of the three leading cloud platforms (Amazon Web Services, Microsoft Azure, and Google Cloud Platform) using a selection of representative application workloads from our facility. Our findings show that while current HPC offerings are still nascent, significant progress is being made to address the present shortcomings. At the same time, significant challenges and questions remain about whether HPC cloud offerings will be able to deliver the full range of expected benefits.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Evaluation of pre-training large language models on leadership-class supercomputers

Large language models (LLMs) have arisen rapidly to the center stage of artificial intelligence as the foundation models applicable to many downstream learning tasks. However, how to effectively build, train, and serve such models for many high-stake and first-principle-based scientific use cases are both of great interests and of great challenges. Moreover, pre-training LLMs with billions or even trillions of parameters can be prohibitively expensive not just for academic institutions, but also for well-funded industrial and government labs. Furthermore, the energy cost and the environmental impact of developing LLMs must be kept in mind. Here, in this work, we conduct a first-of-its-kind performance analysis to understand the time and energy cost of pre-training LLMs on the Department of Energy (DOE)’s leadership-class supercomputers. Employing state-of-the-art distributed training techniques, we evaluate the computational performance of various parallelization approaches at scale for a range of model sizes, and establish a projection model for the cost of full training. Our findings provide baseline results, best practices, and heuristics for pre-training such large models that should be valuable to HPC community at large. We also offer insights and optimization strategies for using the first exascale computing system, Frontier, to train models of the size of GPT-3 and beyond.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

DeepThermo: Deep Learning Accelerated Parallel Monte Carlo Sampling for Thermodynamics Evaluation of High Entropy Alloys

Since the introduction of Metropolis Monte Carlo (MC) sampling, it and its variants have become standard tools used for thermodynamics evaluations of physical systems. However, a long-standing problem that hinders the effectiveness and efficiency of MC sampling is the lack of a generic method (a.k.a. MC proposal) to update the system configurations. Consequently, current practices are not scalable. Here we propose a parallel MC sampling framework for thermodynamics evaluation—DeepThermo. By using deep learning–based MC proposals that can globally update the system configurations, we show that DeepThermo can effectively evaluate the phase transition behaviors of high entropy alloys, which have an astronomical configuration space. For the first time, we directly evaluate a density of states expanding over a range of ~e 10,000 for a real material. We also demonstrate DeepThermo’s performance and scalability up to 3,000 GPUs on both NVIDIA V100 and AMD MI250X-based supercomputers.

Yin, Junqi↗

DOE OSTI2023

A scalable transformer model for real-time decision making in neutron scattering experiments

The U.S. Department of Energy's (DOE's) neutron research facilities at Oak Ridge National Laboratory (ORNL), including the High Flux Isotope Reactor (HFIR) and the Spallation Neutron Source (SNS), are a state-of-the-art neutron scattering facility that allows researchers to study the structure and dynamics of materials at the atomic scale. At the SNS, neutrons are measured using the time-of-flight (TOF) technique as they move through a neutron beamline to interact with a sample. Large volumes of neutron scattering data are collected and recorded in neutron event mode. Optimal productivity of the TOF instrument is limited due to the lack of real-time data analysis tools. The large amount of data generated by the experiments can be challenging to process and analyze in real time, particularly for experiments that require rapid feedback and adjustment of experimental parameters. The regular computer/workstation cannot keep up with the experiment speed to provide real-time feedback to adjust experimental parameters, so connecting the supercomputers available to the neutron facility is necessary to achieve real-time data analysis and experiment steering. To address this challenge, we exploit the Frontier supercomputer at Oak Ridge Leadership Computing Facility (OLCF) to train a scalable temporal fusion transformer model for real-time decision making of TOF neutron scattering experimentation. Here, in this paper, we present the results using Frontier to provide the processing power needed to rapidly process and analyze large volumes of single-crystal diffraction data collected at TOPAZ, a neutron time-of-flight Laue single-crystal diffractometer at the SNS.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

Toward an Autonomous Workflow for Single Crystal Neutron Diffraction

The operation of the neutron facility relies heavily on beamline scientists. Some experiments can take one or two days with experts making decisions along the way. Leveraging the computing power of HPC platforms and AI advances in image analyses, here we demonstrate an autonomous workflow for the single-crystal neutron diffraction experiments. The workflow consists of three components: an inference service that provides real-time AI segmentation on the image stream from the experiments conducted at the neutron facility, a continuous integration service that launches distributed training jobs on Summit to update the AI model on newly collected images, and a frontend web service to display the AI tagged images to the expert. Ultimately, the feedback can be directly fed to the equipment at the edge in deciding the next-step experiment without requiring an expert in the loop. With the analyses of the requirements and benchmarks of the performance for each component, this effort serves as the first step toward an autonomous workflow for real-time experiment steering at ORNL neutron facilities.

Yin, Junqi↗

DOE OSTI2023

Machine Learning for First Principles Calculations of Material Properties for Ferromagnetic Materials

The investigation of finite temperature properties using Monte-Carlo (MC) methods requires a large number of evaluations of the system’s Hamiltonian to sample the phase space needed to obtain physical observables as function of temperature. DFT calculations can provide accurate evaluations of the energies, but they are too computationally expensive for routine simulations. To circumvent this problem, machine-learning (ML) based surrogate models have been developed and implemented on high-performance computing (HPC) architectures. In this paper, we describe two ML methods (linear mixing model and HydraGNN) as surrogates for first principles density functional theory (DFT) calculations with classical MC simulations. These two surrogate models are used to learn the dependence of target physical properties from complex compositions and interactions of their constituents. We present the predictive performance of these two surrogate models with respect to their complexity while avoiding the danger of overfitting the model. An important aspect of our approach is the periodic retraining with newly generated first principles data based on the progressive exploration of the system’s phase space by the MC simulation. The numerical results show that HydraGNN model attains superior predictive performance compared to the linear mixing model for magnetic alloy materials.

Eisenbach, Markus↗

DOE OSTI2022

Intelligent resolution: Integrating Cryo-EM with AI-driven multi-resolution simulations to observe the severe acute respiratory syndrome coronavirus-2 replication-transcription machinery in action

The severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) replication transcription complex (RTC) is a multi-domain protein responsible for replicating and transcribing the viral mRNA inside a human cell. Attacking RTC function with pharmaceutical compounds is a pathway to treating COVID-19. Conventional tools, e.g., cryo-electron microscopy and all-atom molecular dynamics (AAMD), do not provide sufficiently high resolution or timescale to capture important dynamics of this molecular machine. Consequently, we develop an innovative workflow that bridges the gap between these resolutions, using mesoscale fluctuating finite element analysis (FFEA) continuum simulations and a hierarchy of AI-methods that continually learn and infer features for maintaining consistency between AAMD and FFEA simulations. We leverage a multi-site distributed workflow manager to orchestrate AI, FFEA, and AAMD jobs, providing optimal resource utilization across HPC centers. Our study provides unprecedented access to study the SARS-CoV-2 RTC machinery, while providing general capability for AI-enabled multi-resolution simulations at scale.

Trifan, Anda↗

DOE OSTI2022

Language models for the prediction of SARS-CoV-2 inhibitors

The COVID-19 pandemic highlights the need for computational tools to automate and accelerate drug design for novel protein targets. We leverage deep learning language models to generate and score drug candidates based on predicted protein binding affinity. We pre-trained a deep learning language model (BERT) on ∼9.6 billion molecules and achieved peak performance of 603 petaflops in mixed precision. Our work reduces pre-training time from days to hours, compared to previous efforts with this architecture, while also increasing the dataset size by nearly an order of magnitude. For scoring, we fine-tuned the language model using an assembled set of thousands of protein targets with binding affinity data and searched for inhibitors of specific protein targets, SARS-CoV-2 Mpro and PLpro. We utilized a genetic algorithm approach for finding optimal candidates using the generation and scoring capabilities of the language model. Our generalizable models accelerate the identification of inhibitors for emerging therapeutic targets.

Blanchard, Andrew E.↗

DOE OSTI2022

Stable parallel training of Wasserstein conditional generative adversarial neural networks

In this work, we propose a stable, parallel approach to train Wasserstein conditional generative adversarial neural networks (W-CGANs) under the constraint of a fixed computational budget. Differently from previous distributed GANs training techniques, our approach avoids inter-process communications, reduces the risk of mode collapse and enhances scalability by using multiple generators, each one of them concurrently trained on a single data label. The use of the Wasserstein metric also reduces the risk of cycling by stabilizing the training of each generator. We illustrate the approach on the CIFAR10, CIFAR100, and ImageNet1k datasets, three standard benchmark image datasets, maintaining the original resolution of the images for each dataset. Performance is assessed in terms of scalability and final accuracy within a limited fixed computational time and computational resources. To measure accuracy, we use the inception score, the Fréchet inception distance, and image quality. An improvement in inception score and Fréchet inception distance is shown in comparison to previous results obtained by performing the parallel approach on deep convolutional conditional generative adversarial neural networks as well as an improvement of image quality of the new images created by the GANs approach. Weak scaling is attained on both datasets using up to 2000 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Learning to Scale the Summit: AI for Science on a Leadership Supercomputer

The Summit system at Oak Ridge National Lab-oratory (ORNL) has been the world's top AI for science su-percomputer for several years, ranked world's fastest computer at its 2018 launch and currently top system in the US and #2 on the TOP5OO list. Summit's purposeful design to handle both conventional modeling and simulation science and emerging AI workloads has made it a leading destination for AI-powered computational science. We report here on AI for science usage on Summit near the midpoint of its lifespan. We review AI usage across the many science projects that have used Summit. We then examine in detail a set of applications scaling AI to full system as well as projects implementing AI-coordinated science discovery workflows on Summit. Finally, we offer some observations regarding the future of advancing scientific knowledge and understanding via AI, especially in the context of leadership-class scientific computing.

Joubert, Wayne↗

DOE OSTI2022

Strategies for Integrating Deep Learning Surrogate Models with HPC Simulation Applications

The emerging trend of the convergence of high performance computing (HPC), machine learning/deep learning (ML/DL), and big data analytics presents a host of challenges for large-scale computing campaigns that seek best practices to interleave traditional scientific simulation-based workloads with ML/DL models. A portfolio of systematic approaches to incorporate deep learning into modeling and simulation serves a vital need when we support AI for science at a computing facility. In this paper, we evaluate several strategies for deploying deep learning surrogate models in a representative physics application on supercomputers at the Oak Ridge Leadership Computing Facility (OLCF). We discuss a set of recommended deployment architectures and implementation approaches. We analyze and evaluate these alternatives and show their performance and scalability up to 1000 GPUs on two mainstream platforms equipped with different deep learning hardware and software stacks.

Yin, Junqi↗

DOE OSTI2022

US Department of Energy, Office of Science High Performance Computing Facility Operational Assessment 2021: Oak Ridge Leadership Computing Facility

Oak Ridge National Laboratory’s (ORNL’s) Leadership Computing Facility (OLCF) continues to surpass its operational target goals of supporting users; delivering fast, reliable computational ecosystems; creating innovative solutions for high-performance computing (HPC) needs; contributing to the community to build the next generation HPC workforce, and managing risks, safety, and security associated with operating some of the most powerful computers in the world. The results can be seen in the cutting-edge science conducted by users and the praise from the research community. Calendar year (CY) 2021 saw continued excellence in research supported by the OLCF’s leadership-class computing resources, including Summit (the nation’s most powerful supercomputer), the global scratch file system Alpine, the Scalable Protected Infrastructure (SPI), the Exploratory Visualization Environment for Research in Science and Technology (EVEREST), and the archival mass-storage resource High-Performance Storage System (HPSS). While maintaining access and exceptional user support for Summit, the OLCF continued to make progress on the installation and deployment of Frontier, which will be the nation’s first exascale system when it comes online at the start of CY 2023. Users have already begun running and optimizing scientific codes on Crusher, the OLCF test and development system equipped with Frontier’s architecture. Throughout the year, the OLCF maintained a strong culture of operational excellence, including risk management, workplace safety, and cybersecurity. The OLCF’s rigorous risk management strategy anticipated and mitigated risks, and at this time there are no high-priority operational risks. Similarly, ORNL and the OLCF were committed to operating under the US Department of Energy’s (DOE’s) safety regulations that ensure a safe workplace. Technical staff tracked and monitored existing threats and vulnerabilities within the OLCF while continually developing tools and practices to enhance operations without increasing the facility’s risk. CY 2021 was filled with outstanding results and accomplishments, including a very high rating from users on overall satisfaction for the eighth consecutive year; a tremendous number of node hours delivered to 1,671 researchers on Summit; and the successful delivery of the allocation split of roughly 60%, 20%, and 20% of core-hours offered for the Innovative and Novel Computational Impact on Theory and Experiment (INCITE), Advanced Scientific Computing Research Leadership Computing Challenge (ALCC), and Director’s Discretionary (DD) programs, respectively (Section 2). COVID-19 research remained a focus in 2021, and the ALCC and DD programs allocated over 1 million Summit hours to the COVID-19 High Performance Computing Consortium. These accomplishments, coupled with the high utilization rates (i.e., overall and capability usage), represent the fulfillment of the promise of leadership class machines: efficient facilitation of leadership-class computational applications.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Accelerating Collective Communication in Data Parallel Training across Deep Learning Frameworks

This work develops new techniques within Horovod, a generic communication library supporting data parallel training across deep learning frameworks. In particular, we improve the Horovod control plane by implementing a new coordination scheme that takes advantage of the characteristics of the typical data parallel training paradigm, namely the repeated execution of collectives on the gradients of a fixed set of tensors. Using a caching strategy, we execute Horovod’s existing coordinator-worker logic only once during a typical training run, replacing it with a more efficient decentralized orchestration strategy using the cached data and a global intersection of a bitvector for the remaining training duration. Next, we introduce a feature for end users to explicitly group collective operations, enabling finer grained control over the communication buffer sizes. To evaluate our proposed strategies, we conduct experiments on a world-class supercomputer — Summit. We compare our proposals to Horovod’s original design and observe 2x performance improvement at a scale of 6000 GPUs; we also compare them against tf.distribute and torch.DDP and achieve 12% better and comparable performance, respectively, using up to 1536 GPUs; we compare our solution against BytePS in typical HPC settings and achieve about 20% better performance on a scale of 768 GPUs. Finally, we test our strategies on a scientific application (STEMDL) using up to 27,600 GPUs (the entire Summit) and show that we achieve a near-linear scaling of 0.93 with a sustained performance of 1.54 exaflops (with standard error +- 0.02) in FP16 precision.

Romero, Joshua↗

DOE OSTI2022

Distributed Training and Optimization of Neural Networks

Deep learning models are yielding increasingly better performances thanks to multiple factors. To be successful, model may have large number of parameters or complex architectures and be trained on large dataset. This leads to large requirements on computing resource and turn around time, even more so when hyperparameter optimization is done (e.g. search over model architectures). While this is a challenge that goes beyond particle physics, we review the various ways to do the necessary computations in parallel, and put it in the context of high-energy physics.

Vlimant, Jean-Roch↗

DOE OSTI2022

Anderson Acceleration for Distributed Training of Deep Learning Models

Anderson acceleration (AA) is an extrapolation technique that has recently gained interest in the deep learning (DL) community to speed-up the sequential training of DL models. However, when performed at large scale, the DL training is exposed to a higher risk of getting trapped into steep local minima of the training loss function, and standard AA does not provide sufficient acceleration to escape from these steep local minima. This results in poor generalizability and makes AA ineffective. To restore AA’s advantage to speed-up the training of DL models on large scale computing platforms, we combine AA with an adaptive moving average procedure that boosts the training to escape from steep local minima. By monitoring the relative standard deviation between consecutive iterations, we also introduce a criterion to automatically assess whether the moving average is needed. We applied the method to the following DL instantiations for image classification: (i) ResNet50 trained on the open-source CIFAR100 dataset and (ii) ResNet50 trained on the open-source ImageNet1k dataset. Numerical results obtained using up to 1,536 NVIDIA V100 GPUs on the OLCF supercomputer Summit showed the stabilizing effect of the moving average on AA for all the problems above.

Lupo Pasini, Massimiliano↗

DOE OSTI2021

Stable Parallel Training of Wasserstein Conditional Generative Adversarial Neural Networks : *Full/Regular Research Paper submission for the symposium CSCI-ISAI: Artificial Intelligence

We use a stable parallel approach to train Wasserstein Conditional Generative Adversarial Neural Networks (W-CGANs). The parallel training reduces the risk of mode collapse and enhances scalability by using multiple generators that are concurrently trained, each one of them focusing on a single data label. The use of the Wasserstein metric reduces the risk of cycling by stabilizing the training of each generator. We apply the approach on the CIFAR10 and the CIFAR100 datasets, two standard benchmark datasets with images of the same resolution, but different number of classes. Performance is assessed using the inception score, the Fréchet inception distance, and image quality. An improvement in inception score and Fréchet inception distance is shown in comparison to previous results obtained by performing the parallel approach on deep convolutional conditional generative adversarial neural networks (DC-CGANs). Weak scaling is attained on both datasets using up to 100 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

Lupo Pasini, Massimiliano↗

DOE OSTI2021

Mitigating Catastrophic Forgetting in Deep Learning in a Streaming Setting Using Historical Summary

Recent advancements in scientific equipment and the adaptation of electronics and the Internet of Things (IoT) in our everyday lives resulted in large and complex data production at a high rate. Making meaningful and timely knowledge discovery at a modest cost from this big data is difficult for computing power and storage limitations. Training deep learning models incrementally in a streaming setting can help us with overcoming these limitations. However, in a well-known phenomenon named catastrophic forgetting, incrementally trained models increasingly perform poorly on the past data. To mitigate catastrophic forgetting in training in a streaming setting, we propose constructing a historical summary over time and use the summary with newly arrived data during incremental training. We propose various data summarization techniques such as random sampling, micro clustering, coreset computation, and Auto Encoders to counteract catastrophic forgetting. We built a pipeline for incremental training with a historical summary for training deep learning models for streaming data. We demonstrate the effectiveness of historical summary in mitigating catastrophic forgetting using three case studies involving three different deep learning applications: an Artificial Neural Network (ANN) for classification task on MNIST dataset, a language model (RNN-LM) on the WikiText2 dataset, and a Convolutional Neural Network (CNN), ResNet50 to classify the ImageNet dataset. Through the training of the models, we observe that catastrophic forgetting is evident in ANN and CNN but not in an RNN. For the first task, our method recovers up to 47.9% lost accuracy due to catastrophic forgetting. For the third task, the historical summary recovers classification accuracy by up to 25%. For the second task, though there is not proof of catastrophic forgetting, the training performance (PPL) improves by up to 26% with historical summary.

Dash, Sajal↗

DOE OSTI2021

Magnetic properties of CrFeCoNi based high entropy alloys

Monte Carlo simulations are performed on three high entropy alloys: Cr0.25Fe0.25Co0.25Ni0.25, Cr0.2Fe0.2Co0.2Ni0.2Pd0.2, and Cr0.2Mn0.2Fe0.2Co0.2Ni0.2, with exchange interactions extracted from The ab initio Korringa-Kohn-Rostoker method combined with the coherent potential approximation calculations. Using finite size scaling analyses, we estimate the magnetic phase transition temperature for the four component alloy to be 108(2) K, and although the individual critical exponents are different from 3D Heisenberg universality class, the reduced exponent follows Suzuki weak universality. With the additional Palladium component, the transition temperature elevates to about 200 K. In contrast, we find no magnetic order for the five component alloy with Manganese at any finite temperatures.

Yin, Junqi↗