Search NASA⌕ Search

Engineering topics

Shankar, Mallikarjun (Arjun)

Publications and source records attributed to Shankar, Mallikarjun (Arjun).

At least 19 records

OLCF’s Advanced Computing Ecosystem (ACE): FY24 Efforts for the DOE Integrated Research Infrastructure (IRI) Program

This report highlights significant strides made by Oak Ridge National Laboratory’s Oak Ridge Leadership Computing Facility (OLCF) in advancing computational research and infrastructure. Through the Advanced Computing Ecosystem (ACE) strategic initiative, OLCF has been successfully integrated with Doe’s Integrated Research Infrastructure (IRI) program, establishing itself as a critical framework for enhancing scientific computing capabilities across various domains. This report outlines the activities, accomplishments, and future directions of ACE, emphasizing its role in developing cutting-edge technologies, supporting science pilots, and fostering collaborations that drive scientific innovation.

97 MATHEMATICS AND COMPUTING↗

Oak Ridge National Laboratory's Strategic Research and Development Insights for Digital Twins

Oak Ridge National Laboratory (ORNL) is pleased to provide our response to the NITRD RFI on Digital Twins Research and Development. Digital twins are virtual representations of physical systems, leveraging real-time data to simulate and predict behaviors. ORNL is advancing digital twin technology across various disciplines, including neutron scattering, networking, science ecosystems, supercomputing, secure facilities, mobility technologies, materials design and discovery, power systems, fusion reactors, biological sciences, and earth observation. These efforts aim to enhance scientific research, operational efficiency, and decision-making processes. ORNL facilities, such as the High Flux Isotope Reactor (HFIR), Grid-C, Spallation Neutron Source (SNS), and Oak Ridge Leadership Computing Facility (OLCF), provide the infrastructure to develop and demonstrate these digital twin technologies. In this document, we lay out key challenges, research gaps, and future opportunities based on our experience with digital twins that aim to serve as useful contributions towards a National Digital Twins R&D Strategic Plan. In the remaining document, we address nine of the thirteen topic areas specified in the RFI.

97 MATHEMATICS AND COMPUTING↗

US Department of Energy, Office of Science, High-Performance Computing Facility: 2023 Operational Assessment Oak Ridge Leadership Computing Facility

The Oak Ridge Leadership Computing Facility (OLCF) was established to accelerate scientific discovery by providing world-leading computational performance and advanced data infrastructure. As a US Department of Energy (DOE) Office of Science user facility, the OLCF has managed the successful deployment and operation of a succession of leadership-class resources dedicated to open science. In addition to these resources, the OLCF staff continually strive to develop innovative processes and technologies, improve security, and empower users through effective allocation management and comprehensive user support and training. These efforts support the advancement of science by the OLCF users and benefit high-performance computing (HPC) facilities around the world. In calendar year (CY) 2023, the OLCF supported 1,676 users and 598 projects and exceeded all targets for user satisfaction. The facility received an average satisfaction score of 4.52 out of 5 on the annual user survey, and 94% of respondents reported a high satisfaction rate with the OLCF overall. Of the 3,619 user tickets submitted in CY 2023, OLCF staff resolved 97% within 3 business days. The facility opened Frontier to full scientific operations this year. Two projects conducted on Frontier received the Association for Computing Machinery (ACM) Gordon Bell Prize and the Gordon Bell Special Prize for Climate Modeling, and a third earned a nomination as a Gordon Bell Prize finalist. The facility’s previous flagship machine, Summit, gained new life and was extended through 2024 in part to help provide resources to the Integrated Research Infrastructure (IRI) projects and the National Artificial Intelligence Research Resource (NAIRR) pilot program. The facility instantiated an Advanced Computing Ecosystem testbed in part to support IRI workflows. OLCF made interactivity easier and more accessible to users than ever through tools like Jupyter notebooks and workflows.

97 MATHEMATICS AND COMPUTING↗

FORGE: Pre-Training Open Foundation Models for Science

Large language models (LLMs) are poised to revolutionize the way we conduct scientific research. However, both model complexity and pre-training cost are impeding effective adoption for the wider science community. Identifying suitable scientific use cases, finding the optimal balance between model and data sizes, and scaling up model training are among the most pressing issues that need to be addressed. In this study, we provide practical solutions for building and using LLM-based foundation models targeting scientific research use cases. We present an end-to-end examination of the effectiveness of LLMs in scientific research, including their scaling behavior and computational requirements on Frontier, the first Exascale supercomputer. We have also developed for release to the scientific community a suite of open foundation models called FORGE with up to 26B parameters using 257B tokens from over 200M scientific articles, with performance either on par or superior to other state-of-the-art comparable models. We have demonstrated the use and effectiveness of FORGE on scientific downstream tasks. Our research establishes best practices that can be applied across various fields to take advantage of LLMs for scientific discovery.

Yin, Junqi↗

Tokenized Data for FORGE Foundation Models

This dataset comprises a vast corpus of 257 billion tokens, accompanied by the corresponding vocabulary file employed in the pre-training of FORGE foundation models. The primary data source for this corpus is scientific documents derived from diverse origins, and they have been tokenized using the Hugging Face BPE tokenizer. Further details about this research can be found in the publication titled FORGE: Pre-Training Open Foundation Models for Science authored by Junqi Yin, Sajal Dash, Feiyi Wang, and Mallikarjun (Arjun) Shankar, presented at SC'23. The data tokenization pipeline and resulting artifacts use CORE data [Ref: Knoth, P., and Zdrahal, Z. (2012). CORE: three access levels to underpin open access. D-Lib Magazine, 18(11/12)]. For use of these data sets for any purpose, please follow the guidelines provided in https://core.ac.uk/terms .

Yin, Junqi↗

High-performance computing leadership to enable advances in artificial intelligence and a thriving compute ecosystem

The past three decades have witnessed the widespread adoption of high-performance computing (HPC) as an essential tool in the advancement of science. From accelerating critical research in the wake of a global pandemic to climate modelling and national security, HPC has become integral to the most cutting-edge scientific research around the world and across application domains. The global competition to debut the next fastest supercomputer keeps pushing the field forward. Meanwhile, increasing capabilities are bringing hallmarks of science fiction, such as artificial intelligence (AI), into daily life. However, the tremendous power of new computing systems comes with an increased concern about equitable access to these resources and their impact on supporting a thriving workforce.

Tourassi, Georgia↗

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↗

Benchmarking for AI for Science

AI has been instrumental for recent developments in a number of domains of the sciences. With several hundred machine learning (ML) algorithms and models, and numerous AI-specific hardware platforms, a common quest for all scientists working on AI for Science is around the selection of machine learning algorithm(s) to solve their domain-specific scientific problems. A number of different initiatives around AI Benchmarking have been set up and have been useful in understanding the benefits of different ML algorithms for different tasks.However, with the majority of these AI Benchmarking initiatives focusing on the conventional notions of benchmarking, where the focus is purely runtime performance (such as training time or inference time), their suitability for benchmarking different ML algorithms for solving scientific problems has been viewed as a performance problem even though both are hardly the same. To make reasonable, explainable, and justifiable advancements in science using AI, it is critical to focus on the merits of these algorithms in handling different domain science problems. In other words, more emphasis must be given on Benchmarking for AI for Science than AI Benchmarking. The vision of the former is not only to assess the performance of ML algorithms, but also to assess, and understand the benefits and merits of different ML algorithms in handling scientific problems. Benchmarking for AI for Science, instead of pure performance focused AI Benchmarking, has several benefits: (i) it has the potential to offer advances in the sciences, much more rapidly than through pure performance-based AI methods, (ii) it will encourage the community to focus on developing better domain-specific AI techniques, particularly given the provision for being able to benchmark different techniques, and (iii) it will encourage hardware manufacturers to focus on developing science-specific hardware subsystems.

Thiyagalingam, Jeyan↗

High Performance Computing Facility Operational Assessment 2022: Oak Ridge Leadership Computing Facility

The Oak Ridge Leadership Computing Facility (OLCF) was established to accelerate scientific discovery by providing world-leading computational performance and advanced data infrastructure. As a US Department of Energy (DOE) Office of Science user facility, the OLCF has managed the successful deployment and operation of a succession of leadership-class resources dedicated to open science. In addition to these resources, the OLCF staff continually strive to develop innovative processes and technologies, improve security, and empower users through allocation management and comprehensive user support and training. These efforts support the advancement of science by the OLCF users and benefit high-performance computing (HPC) facilities around the world. In calendar year (CY) 2022, the OLCF supported 1,681 users and 570 projects and exceeded all targets for user satisfaction. The facility received an average satisfaction score of 4.6 out of 5 on the annual user survey, and 96% of respondents reported a high satisfaction rate with the OLCF overall. Of the 3,212 user tickets submitted in CY 2022, OLCF staff resolved 97% within 3 business days. The facility also introduced several new services for users this year, including weekly virtual office hours with subject matter experts from ORNL and vendor partners; new views in MyOLCF that allow users to analyze allocation and compute usage for a project; the ability to build and run containers on Summit; and improved data visualization support and training resources.

97 MATHEMATICS AND COMPUTING↗

WfBench: Automated Generation of Scientific Workflow Benchmarks

The prevalence of scientific workflows with high computational demands calls for their execution on various distributed computing platforms, including large-scale leadership-class high-performance computing (HPC) clusters. To handle the deployment, monitoring, and optimization of workflow executions, many workflow systems have been developed over the past decade. There is a need for workflow benchmarks that can be used to evaluate the performance of workflow systems on current and future software stacks and hardware platforms.We present a generator of realistic workflow benchmark specifications that can be translated into benchmark code to be executed with current workflow systems. Our approach generates workflow tasks with arbitrary performance characteristics (CPU, memory, and I/O usage) and with realistic task dependency structures based on those seen in production workflows. We present experimental results that show that our approach generates benchmarks that are representative of production workflows, and conduct a case study to demonstrate the use and usefulness of our generated benchmarks to evaluate the performance of workflow systems under different configuration scenarios.

Coleman, Taina↗

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↗

Official Report on the 2021 Computational and Autonomous Workflows Workshop (CAW 2021)

This technical report documents the Computational and Autonomous Workflows (CAW) workshop held at ORNL (Oak Ridge National Laboratory) in July 2021. The theme of the workshop was "FAIR workflows". This document describes the workshop and takeaways for the Department of Energy, for ORNL, and for the Oak Ridge Leadership Computing Facility.

42 ENGINEERING↗

Emerging Trends and Systemic Issues Influencing Today’s U.S. Electric Grid

With increasing shifts from vertically integrated to horizontally structured operations and from centralized to distributed electric power delivery, today’s electric power grid (the “grid”) operators and designers face the challenge of creating an architecture that accommodates a host of diverse requirements. The grid’s modes of operation must address concerns of reliability and stability, new deployments of renewable energy sources, threats from cyber-attacks and natural disasters, and increasingly distributed system operations. Grid modernization calls for a reliable, affordable, sustainable, agile, secure, and resilient grid. However, the modernization of the U.S. power grid is hampered by mounting complexity and diverging objectives from owners and operators and is consequently risky and fraught with potential missteps. Flawed architecture, design, and implementation will lead to stranded investments and lost opportunities. A principled approach to minimize risk and develop a robust grid of the future is to begin with a sound architecture for the grid to inform the design process. Architecture development starts with the context of influencing factors that provide constraints as well as driving goals. This report provides the context of emerging trends and cross-cutting systemic issues in the U.S. electric power grid and serves as a vital input for grid architecture development.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Building an Integrated Ecosystem of Computational and Observational Facilities to Accelerate Scientific Discovery

Future scientific discoveries will rely on flexible ecosystems that incorporate modern scientific instruments, high performance computing resources, parallel distributed data storage, and performant networks across multiple, independent facilities. In addition to connecting physical resources, such an ecosystem presents many challenges in logistics and accessibility, especially in orchestrating computations and experiments that span across leadership computing systems and experimental instruments. Past efforts have typically been application-specific or limited to interfaces for computing resources. This paper proposes a general framework for integrating computation resources and instrument operations, addressing challenges in code development/execution, data staging and collection, software stack, control mechanisms, resource authorization and governance, and hardware integration. We also describe a demonstration use case wherein a Bayesian optimization algorithm running on an edge computing resource guides a scanning probe microscope to autonomously and intelligently characterize a material sample. This science edge ecosystem framework will provide a blueprint for federating multi-institutional, disparate resources and orchestrating scientific workflows across them to enable next-generation discoveries.

Somnath, Suhas↗

Enabling discovery data science through cross-facility workflows

Experimental and observational instruments for scientific research (such as light sources, genome sequencers, accelerators, telescopes and electron microscopes) increasingly require High Performance Computing (HPC) scale capabilities for data analysis and workflow processing. Next-generation instruments are being deployed with higher resolutions and faster data capture rates, creating a big data crunch that cannot be handled by modest institutional computing resources. Often these big data analysis pipelines also require near real-time computing and have higher resilience requirements than the simulation and modeling workloads more traditionally seen at HPC centers. While some facilities have enabled workflows to run at a single HPC facility, there is a growing need to integrate capabilities across HPC facilities to enable cross-facility workflows, either to provide resilience to an experiment, increase analysis throughput capabilities, or to better match a workflow to a particular architecture. In this paper we describe the barriers to executing complex data analysis workflows across HPC facilities and propose an architectural design pattern for enabling scientific discovery using cross-facility workflows that includes orchestration services, application programming interfaces (APIs), data access and co-scheduling.

Antypas, Katerina B.↗

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↗

Comparative evaluation of deep learning workloads for leadership-class systems

Deep learning (DL) workloads and their performance at scale are becoming important factors to consider as we design, develop and deploy next-generation high-performance computing systems. Since DL applications rely heavily on DL frameworks and underlying compute (CPU/GPU) stacks, it is essential to gain a holistic understanding from compute kernels, models, and frameworks of popular DL stacks, and to assess their impact on science-driven, mission-critical applications. At Oak Ridge Leadership Computing Facility (OLCF), we employ a set of micro and macro DL benchmarks established through the Collaboration of Oak Ridge, Argonne, and Livermore (CORAL) to evaluate the AI readiness of our next-generation supercomputers. In this paper, we present our early observations and performance benchmark comparisons between the Nvidia V100 based Summit system with its CUDA stack and an AMD MI100 based testbed system with its ROCm stack. We take a layered perspective on DL benchmarking and point to opportunities for future optimizations in the technologies that we consider.

Yin, Junqi↗

Data optimization for large batch distributed training of deep neural networks

Distributed training in deep learning (DL) is common practice as data and models grow. The current practice for distributed training of deep neural networks faces the challenges of communication bottlenecks when operating at scale, and model accuracy deterioration with an increase in global batch size. Present solutions focus on improving message exchange efficiency as well as implementing techniques to tweak batch sizes and models in the training process. The loss of training accuracy typically happens because the loss function gets trapped in a local minima. We observe that the loss landscape minimization is shaped by both the model and training data and propose a data optimization approach that utilizes machine learning to implicitly smooth out the loss landscape resulting in fewer local minima. Our approach filters out data points which are less important to feature learning, enabling us to speed up the training of models on larger batch sizes to improved accuracy.

Gahlot, Shubhankar↗