Search NASA⌕ Search

SEARCH · Search NASA

Results for “HPC system”

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 271 records · Page 15

"Forward" Projects Boost U.S. Leadership in Advanced Computing and Artificial Intelligence

High-performance computing (HPC) has been an indispensable research tool for accessing physical realms difficult, or impossible, achieve with experiment alone. For several decades, the Department of Energy’s (DOE’s) Office of Science has deployed sophisticated HPC systems for solving the nation’s most pressing grand challenge problems in energy, climate change, and human health. In addition, DOE’s National Nuclear Security Administration (NNSA) has adeptly applied HPC in support of key national security objectives, such as nuclear science and stockpile modernization and stewardship. Over time, HPC systems have become increasingly more complex and capable, and as each new machine has come online, scientists and engineers have taken advantage of vast increases in compute power to accelerate scientific discoveries and engineering innovation.

42 ENGINEERING↗

Spatio-Temporal Analysis of HPC I/O and Connection Data

The HPC system consists of a set of layers of software and hardware for I/O and networking. System logs are helpful resources to understand what is going on in the system. A challenge is that it is non-trivial to analyze the logs maintained in various levels of the stack. Independent analysis might lead to an incomplete conclusion due to the limited coverage of each log. This work takes a comprehensive approach to analysis that incorporates the logs in the multiple layers and components, in order to facilitate the detection of anomalous activities. This research aims to identify and predict potential performance bottlenecks in the HPC system, by capturing the temporal variation patterns from heterogeneous, high-dimensional, and non-linear log data. In this paper, we share our preliminary efforts for spatio-temporal analysis of HPC I/O and connection data, with our initial observations from the analysis of one-week HPC log data sets collected from one of NERSC systems.

Choi, Jinhwan↗

Gnawts

High Performance Computing (HPC) systems generate millions of inter-related event log lines per day from thousands of computers, providing feedback on both successful and unsuccessful operations within the system. Thorough analysis of these logs is often required for root cause analysis of problems and tuning for greatest performance. This "app" for the "Gnawts" log analysis software (http://www.gnawts.com) facilitates faster and deeper insights into HPC system logs.

Stearley, Jon↗

Gnawts

High Performance Computing (HPC) systems generate millions of inter-related event log lines per day from thousands of computers, providing feedback on both successful and unsuccessful operations within the system. Thorough analysis of these logs is often required for root cause analysis of problems and tuning for greatest performance. This "app" for the "Gnawts" log analysis software (http://www.gnawts.com) facilitates faster and deeper insights into HPC system logs.

Jon Stearley, Charles Snider↗

Performance Characterization of De Novo Genome Assembly on Leading Parallel Systems.

De novo genome assembly is one of the most important and challenging computational problems in modern genomics; further, it shares algorithms and communication patterns important to other graph analytic and irregular applications. Unlike simulations, it has no floating point arithmetic and is dominated by small memory transactions within and between computing nodes. In this work, we focus on the highly scalable HipMer assembler and identify the dominant algorithms and communication patterns, also using microbenchmarks to capture the workload. We evaluate HipMer on a variety of platforms from the latest HPC systems to ethernet clusters. HipMer performs well on all single node systems, including the Xeon Phi manycore architecture. Given large enough problems, it also demonstrates excellent scaling across nodes in an HPC system, but requires a high speed network with low overhead and high injection rates. Our results shed light on the architectural features that are most important for achieving good parallel efficiency on this and related problems.

Ellis, Marquita↗

Strategies for Fault Tolerance in Multicomponent Applications

This paper discusses on-going work with the Integrated Plasma Simulator (IPS), a framework for coupled multiphysics simulations of plasmas, to allow simulations to run through the loss of nodes on which the simulation is executing. While many different techniques are available to improve the fault tolerance of computational science applications on high-performance computer systems, checkpoint/restart (C/R) remains virtually the only one that see widespread use in practice. Our focus here is to augment the traditional C/R approach with additional techniques that can provide a more localized and tailored response to faults based on the ability to restart failed tasks on an individual basis, and the use of information external to the application itself in order to guide decision-making, in many cases avoiding the need to stop and restart the entire simulation. This capability involves several features within the IPS framework, and leverages the Fault Tolerance Backplane, a publish/subscribe event service to disseminate fault-related information throughout HPC systems, to obtain information from the Reliability, Availability and Serviceability (RAS) subsystem of the HPC system. This work is described in the context of Cray XT-series computer systems for concreteness, but is applicable to other environments as well. As part of the analysis of this work, we discuss the requirements to generalize this approach to other complex simulation applications beyond the Integrated Plasma Simulator.

99 GENERAL AND MISCELLANEOUS↗

2013 R&D 100 Award: ‘Miniapps’ Bolster High Performance Computing

Two Livermore computer scientists served on a Sandia National Laboratories-led team that developed Mantevo Suite 1.0, the first integrated suite of small software programs, also called "miniapps," to be made available to the high performance computing (HPC) community. These miniapps facilitate the development of new HPC systems and the applications that run on them. Miniapps (miniature applications) serve as stripped down surrogates for complex, full-scale applications that can require a great deal of time and effort to port to a new HPC system because they often consist of hundreds of thousands of lines of code. The miniapps are a prototype that contains some or all of the essentials of the real application but with many fewer lines of code, making the miniapp more versatile for experimentation. This allows researchers to more rapidly explore options and optimize system design, greatly improving the chances the full-scale application will perform successfully. These miniapps have become essential tools for exploring complex design spaces because they can reliably predict the performance of full applications.

97 MATHEMATICS AND COMPUTING↗

Online data analysis and reduction: An important co-design motif for extreme-scale computers

A growing disparity between supercomputer computation speeds and I/O rates means that it is rapidly becoming infeasible to analyze supercomputer application output only after that output has been written to a file system. Instead, data-generating applications must run concurrently with data reduction and/or analysis operations, with which they exchange information via high-speed methods such as interprocess communications. The resulting parallel computing motif, online data analysis and reduction (ODAR), has important implications for both application and HPC systems design. Here we introduce the ODAR motif and its co-design concerns, describe a co-design process for identifying and addressing those concerns, present tools that assist in the co-design process, and present case studies to illustrate the use of the process and tools in practical settings.

Data Analysis↗

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↗

Quantum/AI Topology-Aware Latency-Adaptive HPC Workflow Scheduling Optimization

The growing demand for more powerful high-performance computing (HPC) systems has led to a steady rise in energy consumption by supercomputing worldwide. This study is focused on comparing our Application-Topology Mapper (ATMapper) to the popular Simple Linux Utility for Resource Management (SLURM) for the purpose of exploring methods that can further optimize job-scheduling within HPC systems. ATMapper is an Artificial-Intelligence based approach to job-scheduling that is currently being enhanced with quantum annealing (QA) to generate optimal schedules faster. We are applying QA to speedup our ATMapper process to achieve higher computing efficiency, thereby reducing HPC energy consumption. Here, we examine how four job-scheduling approaches perform in processor node assignment when using an example network architecture of 4 interconnected nodes. Using a specialized script, we are assessing the schedule of a computation flow with 11 interdependent tasks. The data movements among nodes were tracked to count for the number of interactions (network hops) between nodes needed to complete the tasks. The total number of hops and the job completion time were then used to quantify the efficiency of the different mapping approaches. In addition to SLURM, we also compare our ATMapper to the QA-enabled LBNL TIGER and the D-Wave Distributed Computing processor assignment approaches. The preliminary results showed that our topology-aware, latency-adaptive ATMapper is significantly more efficient when compared to the other scheduling approaches due to its load-imbalance network allocation. The scheduler displayed a computing efficiency of 53% by performing significantly fewer network hops than its alternatives. By reducing the number of hops, ATMapper was able to perform all 11 tasks by using only 3 nodes out of given 4. This research indicates the potential to use QA/AI for HPC job-scheduling. Later, we will test a SLURM simulator program to draw further comparisons on the effectiveness of ATMapper's scheduling approach. The results of this comparison will serve as a baseline for later improving SLURM's performance using a QA-enhanced ATMapper approach.

Caraveo, Braulio [University of Huston - Clear Lak↗

High-Throughput Computing: Case Study of Medical Image Processing Applications

HPC is designed for large-scale simulations using monolithic codes of tightly coupled processes highly optimized to deliver decreased time to solution. Medical image processing is not a traditional field of HPC. Similar to AI applications, medical image processing parses large datasets, typically multiple times, to support a variety of studies for classification, diagnosis or monitoring purposes. The convergence of AI, HPC and Big Data encouraged more fields using image processing to transition to HPC. However, not all applications benefit from the same optimizations. In this paper we focus on high throughput medical image processing applications that analyze a huge dataset of small MRI images and that require HPC systems to decrease the time of parsing the entire dataset and not individual MRIs. We show in this research the performance of running SLANT, an image processing application for a whole brain segmentation, on large-scale systems and highlight performance limitations. We present optimizations prioritizing throughput that exhibit a 3.5x speed-up on the Summit Supercomputer that can be used as a baseline for building a high-throughput execution framework for other HPC systems.

Predescu, Maria↗

Addressing GPU memory limitations for Graph Neural Networks in High-Energy Physics applications

Introduction Reconstructing low-level particle tracks in neutrino physics can address some of the most fundamental questions about the universe. However, processing petabytes of raw data using deep learning techniques poses a challenging problem in the field of High Energy Physics (HEP). In the Exa.TrkX Project, an illustrative HEP application, preprocessed simulation data is fed into a state-of-art Graph Neural Network (GNN) model, accelerated by GPUs. However, limited GPU memory often leads to Out-of-Memory (OOM) exceptions during training, due to the large size of models and datasets. This problem is exacerbated when deploying models on High-Performance Computing (HPC) systems designed for large-scale applications. Methods We observe a high workload imbalance issue during GNN model training caused by the irregular sizes of input graph samples in HEP datasets, contributing to OOM exceptions. We aim to scale GNNs on HPC systems, by prioritizing workload balance in graph inputs while maintaining model accuracy. Our paper introduces diverse balancing strategies aimed at decreasing the maximum GPU memory footprint and avoiding the OOM exception, across various datasets. Results Our experiments showcase memory reduction of up to 32.14% compared to the baseline. We also demonstrate the proposed strategies can avoid OOM in application. Additionally, we create a distributed multi-GPU implementation using these samplers to demonstrate the scalability of these techniques on the HEP dataset. Discussion By assessing the performance of these strategies as data loading samplers across multiple datasets, we can gauge their effectiveness in both single-GPU and distributed environments. Our experiments, conducted on datasets of varying sizes and across multiple GPUs, broaden the applicability of our work to various GNN applications that handle input datasets with irregular graph sizes.

Lee, Claire Songhyun↗

SpackDev: Multi-Package Development with Spack

High Energy Physics (HEP) software environments often involve ~ hundreds of external packages and libraries, and similar numbers of experiment-specific, science-critical packages—many under continuous development. Managing coherent releases of the external software stack is challenging enough, but managing the highly-collaborative—and distributed—development of a large body of code against such a stack adds even more complexity and room for error. Spack is a popular Python-based package management tool with a specific focus on the needs of High Performance Computing (HPC) systems and system administrators whose strength is orchestrating the discrete download, build, testing, and installation of pre-packaged or tagged third-party software against similarly stable dependencies. As such it is becoming increasingly popular within HEP as that community makes increasing use of HPC facilities, and as efforts to develop future HPC systems utilize Spack to provide scientific software on those platforms. SpackDev is a system to facilitate the simultaneous development of interconnected sets of packages. Intended to handle packages without restriction to one internal build system, SpackDev is integrated with Spack as a command extension in order to leverage features such as dependency calculations and build system configuration, and is generally applicable outside HEP. We describe SpackDev’s features and development over the last two years, initial experience using SpackDev in the context of the LArSoft liquid argon detector toolkit, and work remaining before it can be considered a fully-functional multi-package build system for HEP experiments utilizing Spack.

Green, Chris↗

Portable and Transparent Message Compression in MPI Libraries to Improve the Performance and Scalability of Parallel Applications

The goal of this project has been to develop a lossless compression algorithm for message-passing libraries that can accelerate HPC systems by reducing the communication time. Because both compression and decompression have to be performed in software in real time, the algorithm has to be extremely fast while still delivering a good compression ratio. During the first half of this project, they designed a new compression algorithm called FPC for scientific double-precision data, made the source code available on the web, and published two papers describing its operation, the first in the proceedings of the Data Compression Conference and the second in the IEEE Transactions on Computers. At comparable average compression ratios, this algorithm compresses and decompresses 10 to 100 times faster than BZIP2, DFCM, FSD, GZIP, and PLMI on the three architectures tested. With prediction tables that fit into the CPU's L1 data acache, FPC delivers a guaranteed throughput of six gigabits per second on a 1.6 GHz Itanium 2 system. The C source code and documentation of FPC are posted on-line and have already been downloaded hundreds of times. To evaluate FPC, they gathered 13 real-world scientific datasets from around the globe, including satellite data, crash-simulation data, and messages from HPC systems. Based on the large number of requests they received, they also made these datasets available to the community (with permission of the original sources). While FPC represents a great step forward, it soon became clear that its throughput was too slow for the emerging 10 gigabits per second networks. Hence, no speedup can be gained by including this algorithm in an MPI library. They therefore changed the aim of the second half of the project. Instead of implementing FPC in an MPI library, they refocused their efforts to develop a parallel compression algorithm to further boost the throughput. After all, all modern high-end microprocessors contain multiple CPUs on a single die and will undoubtedly contain even larger numbers of CPUs in the future. Hence, a compute node in a parallel system is likely to have multiple processors per interconnection link. Thus, they investigated how to best exploit multiple CPUs to compress and decompress data for a single link, i.e., they developed a parallel version of the FPC algorithm called pFPC. On a 4-core 3 GHz Xeon system, pFPC exceeds 10 gigabits per second both in compression and decompression throughput while still reaching a compression ratio of 18%, which is reasonable for their double-precision floating-point datasets. At half this throughput, it achieves a 41% compression ratio. Moreover, the algorithm scales well to four cores and promises even higher throughputs on larger machines. Additional results and a description of the pFPC algorithm have recently been published in the proceedings of the Data Compression Conference. The C source code of pFPC is also publicly available.

97 MATHEMATICS AND COMPUTING↗

New Platform Provides Innovation

Sky Bridge is the primary unclassified scientific computing platform at Sandia, serving all Mission Areas within the Laboratories. At the formal ribbon cutting in December 2014, Laboratories Director Paul Hommert emphasized the importance of the Institutional Computing Program in providing a critical core Laboratories capability. The deployment of Sky Bridge more than doubled the computing cycles available in Sandia’s restricted computing environment, while incorporating the same underlying architecture as the existing ASC/TLCC2 compute clusters, thus allowing many existing computer codes to run on the system without any major modifications. Sky Bridge immediately addressed the shortfall in projected computing cycles for Sandia’s programs of energy and environment, nuclear weapons life expectancy program (NW LEP), and Applied Laboratory Technology (ALT). From a technological perspective, Sky Bridge is contributing to the readiness of the data center facility and operations personnel for a transition to liquid cooling in future HPC systems and other high-density systems. By comparing the performance and energy costs of Sky Bridge to similar systems (e.g., Dark Bridge) that are air-cooled rather than liquid-cooled, it will also provide valuable data for making return on investment decisions for future purchases. Installation of Sky Bridge enables decommissioning of the Red Sky supercomputer, reduce energy and support costs, and make room for future HPC systems already planned through the ASC and Institutional Computing programs. Sky Bridge’s use of liquid cooling (cold plate to the systems processors) and more efficient power distribution (480 V direct to the rack) saved over $\$$700, 000 in construction costs while reducing annual operating costs by $\$$120,000.

97 MATHEMATICS AND COMPUTING↗

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↗

Divide and Conquer (DC) BLAST: fast and easy BLAST execution within HPC environments

Bioinformatics is currently faced with very large-scale data sets that lead to computational jobs, especially sequence similarity searches, that can take absurdly long times to run. For example, the National Center for Biotechnology Information (NCBI) Basic Local Alignment Search Tool (BLAST and BLAST+) suite, which is by far the most widely used tool for rapid similarity searching among nucleic acid or amino acid sequences, is highly central processing unit (CPU) intensive. While the BLAST suite of programs perform searches very rapidly, they have the potential to be accelerated. In recent years, distributed computing environments have become more widely accessible and used due to the increasing availability of high-performance computing (HPC) systems. Therefore, simple solutions for data parallelization are needed to expedite BLAST and other sequence analysis tools. However, existing software for parallel sequence similarity searches often requires extensive computational experience and skill on the part of the user. In order to accelerate BLAST and other sequence analysis tools, Divide and Conquer BLAST (DCBLAST) was developed to perform NCBI BLAST searches within a cluster, grid, or HPC environment by using a query sequence distribution approach. Scaling from one (1) to 256 CPU cores resulted in significant improvements in processing speed. Thus, DCBLAST dramatically accelerates the execution of BLAST searches using a simple, accessible, robust, and parallel approach. DCBLAST works across multiple nodes automatically and it overcomes the speed limitation of single-node BLAST programs. DCBLAST can be used on any HPC system, can take advantage of hundreds of nodes, and has no output limitations. Thus, this freely available tool simplifies distributed computation pipelines to facilitate the rapid discovery of sequence similarities between very large data sets.

59 BASIC BIOLOGICAL SCIENCES↗

HPC Digital Twins for Evaluating Scheduling Policies, Incentive Structures and their Impact on Power and Cooling

Schedulers are critical for optimal resource utilization in high-performance computing. Traditional methods to evaluate sched- ulers are limited to post-deployment analysis, or simulators, which do not model associated infrastructure. In this work, we present the first-of-its-kind integration of scheduling and digital twins in HPC. This enables what-if studies to understand the impact of parameter configurations and scheduling decisions on the physical assets, even before deployment, or regarching changes not easily realizable in production. We (1) provide the first digital twin framework extended with scheduling capabilities, (2) integrate various top-tier HPC systems given their publicly available datasets, (3) implement extensions to integrate external scheduling simulators. Finally, we show how to (4) implement and evaluate incentive structures, as- well-as (5) evaluate machine learning based scheduling, in such novel digital-twin based meta-framework to prototype scheduling. Our work enables what-if scenarios of HPC systems to evaluate sustainability, and the impact on the simulated system.

Maiterth, Matthias [ORNL] (ORCID:000000018698460X)↗