Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “High Performance Computing (HPC)”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,279 records · Page 71

Union: An Automatic Workload Manager for Accelerating Network Simulation

With the rapid growth of the machine learning applications, the workloads of future HPC systems are anticipated to be a mix of scientific simulation, big data analytics, and machine learning applications. Simulation is a great research vehicle to understand the performance implications of co-running scientific applications with big data and machine learning workloads on large-scale systems. In this paper, we present Union, a workload manager that provides an automatic framework to facilitate hybrid workload simulation in CODES. Furthermore, we use Union, along with CODES, to investigate various hybrid workloads composed of traditional simulation applications and emerging learning applications on two dragonfly systems. The experiment results show that both message latency and communication time are important performance metrics to evaluate network interference. Network interference on HPC applications is more reflected by the message latency variation, whereas ML application performance depends more on the communication time.

heterogeneous workloads↗

PIPER: Performance Insights for Programmers and Exascale Runtimes

During the lifetime of the project, the PIPER at the University of Utah worked in collaboration with the team from the partner institutions to establish a set of software tools for analyzing and improving the performance parallel computing tools. These activities complement and complete the work performed by the other X-Stack 2 projects with the definition of a common data models and computing infrastructures, which have been a central element in the targeted performance analysis pipeline and new programming models. The University of Utah team have focused mainly on developing data analysis and visualization systems to facilitate exploration, comparison and evaluation of the behavior and performance of complex algorithms in modern HPC infrastructures. This is involved several demonstrations of the performance achieved and presentation of the results in a way that facilitates the users understand bottlenecks and ways to overcome them. One major focus of the University of Utah activities has been the analysis of the performance characteristics of Fat-Tree and Dragonfly topology networks. The work involved the development of practical visualization systems especially in collaboration with the team at LLNL, and published several papers and gave invited talks at prestigious venues such as the Salishan Conference on High-Speed Computing or the International Conference on High Performance Computing, Data, and Analytics. In addition, the university of Utah team has advanced the core activities on analyzing the performance characteristics of I/O and other data movement systems and their practical use for scaling Uintah and other simulation codes poised for use in the exascale computing environments of the future.

42 ENGINEERING↗

Auto-HPCnet: An Automatic Framework to Build Neural Network-based Surrogate for High-Performance Computing Applications

High-performance computing communities are increasingly adopt- ing Neural Networks (NN) as surrogate models in their applications to generate scientific insights. Replacing an execution phase in the application with NN models can bring significant performance im- provement. However, there is a lack of tools that can help domain scientists automatically apply NN-based surrogate models to HPC applications. We introduce a framework, named Auto-HPCnet, to democratize the usage of NN-based surrogates. Auto-HPCnet is the first end-to-end framework that makes past proposals for the NN-based surrogate model practical and disciplined. Auto-HPCnet introduces a workflow to address unique challenges when apply- ing the approximation, such as feature acquisition and meeting the application-specific constraint on the quality of final computation outcome. We show that Auto-HPCnet can leverage NN for a set of HPC applications and achieve 5.50× speedup on average (up to 16.8× speedup and with data preparation cost included) while meeting the application-specific constraint on the final computation quality.

Dong, Wenqian↗

ZFP Hardware Implementation

ZFP is a floating-point compression format gaining traction in high-performance computing applications. A software implementation has demonstrated the ability to reduce data movement across communication channels and to reduce the footprint of floating-point arrays in memory. Nevertheless, any benefit in performance is limited to the spare compute cycles available before reaching bandwidth limitations. A hardware implementation of ZFP has the potential to substantially improve HPC application performance. Hardware compression speedup over software alone on a single CPU core ranges from 15x for 1-D arrays to over 200x for 3-D arrays. The implementation of ZFP is sourced in SystemC to facilitate its evaluation in various architectures. The encode pipeline consists of several modules. Uncompressed blocks of floating-point numbers in IEEE format flow into the pipeline in a stream and a compressed bitstream flows out from the pipeline. The decode pipeline is the inverse of the encode pipeline. A modular design enables formats other than IEEE (such as posits) to be considered with minor adaptations. Best performance will be realized from the hardware ZFP unit when batches of blocks are processed at a time. The implementation supports 1-D, 2- D and 3-D blocks of floating-point numbers. The hardware implementation of ZFP has been validated with the software implementation of ZFP. C++ template parameters are used to specify the bit width of floating-point numbers and the array dimension of the encoder. A test bench program has been created with several test cases, some with continuous data that is ideal for ZFP, and others with extreme cases containing numbers near or at the maximum or minimum values supported by the number format.

Lloyd, GS↗

Dynamic Communicators in MPI

This paper describes a proposal to add support for dynamic communicators to the MPI standard. This adds the ability to grow or shrink a specified communicator, under well specified circumstances. The goal is to make it possible for a new class of applications - long-running, mission-critical, loosely coupled applications, running in a highly dynamic environment - to use MPI libraries for their communications needs, and to enable HPC applications to adjust to changing system resources. Implementation analysis indicates that performance impact on existing high-performance MPI applications should be minimal, or non-existent. The changes to MPI implementations are expected to be well compartmentalized.

97 MATHEMATICS AND COMPUTING↗

Confidence: Analyzing Performance With Empirical Probabilities

Variability in the performance of shared system components is a major obstacle in analyzing the effective throughput of leadership class computers. Shared file systems and networks are serious impediments to achieving repeatable application performance on HPC systems. In particular, performance analysts are likely to be interested in quantifying differences between average-case behavior, worst-case behavior, and standard deviation for shared system components. Typical descriptions of these statistics assume a normal distribution; however, in non-linear and multi-modal performance distributions, summary statistics are often misleading. In this paper we describe Confidence, a tool for analyzing the full spectrum of performance for a benchmarking code. By including all of the experimental outcomes in the analysis without discarding any measurements, Confidence enables a novel analysis of benchmark performance. Finally, we use Confidence to characterize the network performance of three high performance computer systems, and we provide recommendations for leveraging network performance in latency sensitive applications.

Settlemyer, Bradley W↗

Scaling SQL to the Supercomputer for Interactive Analysis of Simulation Data

AI and simulation workloads consume and generate large amounts of data that need to be searched, transformed and merged with other data. With the goal of treating data as a first-class citizen inside a traditionally compute-centric HPC environment, we explore how the use of accelerators and high-speed interconnects can speed up tasks which otherwise constitute bottlenecks in computational discovery workflows. BlazingSQL is SQL engine that runs natively on NVIDIA GPUs and supports internode communication for fast analytics on terabyte-scale tabular data sets. We show how a fast interconnect improves query performance if leveraged through the Unified Communication X (UCX) middleware. We envision that future computing platforms will integrate accelerated database query capabilities for immediate and interactive analysis of large simulation data.

Glaser, Jens↗

HPC infrastructure to support the next-generation ARM facility data operations

The Department of Energy's (DOE) Atmospheric Radiation Measurement (ARM) Climate Research Facility is establishing an adaptive data services and operations architecture in support of the Next-Generation ARM Facility as explained in its Decadal Vision. In this paper, we describe the capabilities of the ARM Data Center (ADC) and the upcoming high-performance computing infrastructure in support of this Next-Generation ARM Facility.

Prakash, Giri↗

Steps toward fault-tolerant quantum chemistry.

Developing quantum chemistry programs on the coming generation of exascale computers will be a difficult task. The programs will need to be fault-tolerant and minimize the use of global operations. This work explores the use a task-based model that uses a data-centric approach to allocate work to different processes as it applies to quantum chemistry. After introducing the key problems that appear when trying to parallelize a complicated quantum chemistry method such as coupled-cluster theory, we discuss the implications of that model as it pertains to the computational kernel of a coupled-cluster program - matrix multiplication. Also, we discuss the extensions that would required to build a full coupled-cluster program using the task-based model. Current programming models for high-performance computing are fault-intolerant and use global operations. Those properties are unsustainable as computers scale to millions of CPUs; instead one must recognize that these systems will be hierarchical in structure, prone to constant faults, and global operations will be infeasible. The FAST-OS HARE project is introducing a scale-free computing model to address these issues. This model is hierarchical and fault-tolerant by design, allows for the clean overlap of computation and communication, reducing the network load, does not require checkpointing, and avoids the complexity of many HPC runtimes. Development of an algorithm within this model requires a change in focus from imperative programming to a data-centric approach. Quantum chemistry (QC) algorithms, in particular electronic structure methods, are an ideal test bed for this computing model. These methods describe the distribution of electrons in a molecule, which determine the properties of the molecule. The computational cost of these methods is high, scaling quartically or higher in the size of the molecule, which is why QC applications are major users of HPC resources. The complexity of these algorithms means that MPI alone is insufficient to achieve parallel scaling; QC developers have been forced to use alternative approaches to achieve scalability and would be receptive to radical shifts in the programming paradigm. Initial work in adapting the simplest QC method, Hartree-Fock, to this the new programming model indicates that the approach is beneficial for QC applications. However, the advantages to being able to scale to exascale computers are greatest for the computationally most expensive algorithms; within QC these are the high-accuracy coupled-cluster (CC) methods. Parallel coupledcluster programs are available, however they are based on the conventional MPI paradigm. Much of the effort is spent handling the complicated data dependencies between the various processors, especially as the size of the problem becomes large. The current paradigm will not survive the move to exascale computers. Here we discuss the initial steps toward designing and implementing a CC method within this model. First, we introduce the general concepts behind a CC method, focusing on the aspects that make these methods difficult to parallelize with conventional techniques. Then we outline what is the computational core of the CC method - a matrix multiply - within the task-based approach that the FAST-OS project is designed to take advantage of. Finally we outline the general setup to implement the simplest CC method in this model, linearized CC doubles (LinCC).

99 GENERAL AND MISCELLANEOUS↗

Virtual machine provisioning, code management, and data movement design for the Fermilab HEPCloud Facility

The Fermilab HEPCloud Facility Project has as its goal to extend the current Fermilab facility interface to provide transparent access to disparate resources including commercial and community clouds, grid federations, and HPC centers. This facility enables experiments to perform the full spectrum of computing tasks, including data-intensive simulation and reconstruction. We have evaluated the use of the commercial cloud to provide elasticity to respond to peaks of demand without overprovisioning local resources. Full scale data-intensive workflows have been successfully completed on Amazon Web Services for two High Energy Physics Experiments, CMS and NOνA, at the scale of 58000 simultaneous cores. This paper describes the significant improvements that were made to the virtual machine provisioning system, code caching system, and data movement system to accomplish this work. The virtual image provisioning and contextualization service was extended to multiple AWS regions, and to support experiment-specific data configurations. A prototype Decision Engine was written to determine the optimal availability zone and instance type to run on, minimizing cost and job interruptions. We have deployed a scalable on-demand caching service to deliver code and database information to jobs running on the commercial cloud. It uses the frontiersquid server and CERN VM File System (CVMFS) clients on EC2 instances and utilizes various services provided by AWS to build the infrastructure (stack). We discuss the architecture and load testing benchmarks on the squid servers. We also describe various approaches that were evaluated to transport experimental data to and from the cloud, and the optimal solutions that were used for the bulk of the data transport. Finally, we summarize lessons learned from this scale test, and our future plans to expand and improve the Fermilab HEP Cloud Facility.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Educating HPC Users in the use of advanced computing technology

We examine a multi-modal approach to educating and training users of an advanced computing technology testbed at the Institute for Advanced Computational Science at Stony Brook University. Ookami provides researchers worldwide with access to 176 Fujitsu A64FX compute nodes, this being the same processor technology powering the Japanese Fugaku supercomputer, the fastest computer in the world since June 2020. However, achieving high-performance on this Arm-based, leadership computing technology requires that users be familiar with details of computer architecture, performance analysis and modeling, and high-performance programming models that are commonly omitted in introductory programming courses. Indeed, regardless of their seniority, many of the testbed users are surprisingly unfamiliar with basic concepts such as vectorization, pipelining, latency/bandwidth, roofline models, computing energy/power, threads, and non-uniform memory access. These same concepts also pervade mainstream x86 technologies, so this is of widespread concern. Due to the national/global nature of our user community that is also very diverse in both discipline and experience, the inability to offer formal classes, and our experience that most people do not tend to read online documentation or training materials in sufficient depth, we have consciously employed multiple approaches that heavily emphasize (online) personal interactions and transfer of skills. Online documentation has been organized around best-practices and FAQs; twice-weekly hackathons and office hours via Zoom enable deep dives by both the team and the user community with multiple broad benefits; a Slack channel provides both real time and archived answers and discussions; and workshops, training and webinars target community needs as they arise. Furthermore, the perspective that these tools are being used in an educational setting rather than just for project communication makes them more effective and contributes to community success.

A64FX↗

Sparse Matrix-Based HPC Tomography

Tomographic imaging has benefited from advances in X-ray sources, detectors and optics to enable novel observations in science, engineering and medicine. These advances have come with a dramatic increase of input data in the form of faster frame rates, larger fields of view or higher resolution, so high performance solutions are currently widely used for analysis. Tomographic instruments can vary significantly from one to another, including the hardware employed for reconstruction: from single CPU workstations to large scale hybrid CPU/GPU supercomputers. Furthermore, flexibility on the software interfaces and reconstruction engines are also highly valued to allow for easy development and prototyping. This paper presents a novel software framework for tomographic analysis that tackles all aforementioned requirements. The proposed solution capitalizes on the increased performance of sparse matrix-vector multiplication and exploits multi-CPU and GPU reconstruction over MPI. Furthermore, the solution is implemented in Python and relies on CuPy for fast GPU operators and CUDA kernel integration, and on SciPy for CPU sparse matrix computation. As opposed to previous tomography solutions that are tailor-made for specific use cases or hardware, the proposed software is designed to provide flexible, portable and high-performance operators that can be used for continuous integration at different production environments, but also for prototyping new experimental settings or for algorithmic development. The experimental results demonstrate how our implementation can even outperform state-of-the-art software packages used at advanced X-ray sources worldwide.

97 MATHEMATICS AND COMPUTING↗

MAGMA: Enabling exascale performance with accelerated BLAS and LAPACK for diverse GPU architectures

MAGMA (Matrix Algebra for GPU and Multicore Architectures) is a pivotal open-source library in the landscape of GPU-enabled dense and sparse linear algebra computations. With a repertoire of approximately 750 numerical routines across four precisions, MAGMA is deeply ingrained in the DOE software stack, playing a crucial role in high-performance computing. Notable projects such as ExaConstit, HiOP, MARBL, and STRUMPACK, among others, directly harness the capabilities of MAGMA. In addition, the MAGMA development team has been acknowledged multiple times for contributing to the vendors’ numerical software stacks. Looking back over the time of the Exascale Computing Project (ECP), we highlight how MAGMA has adapted to recent changes in modern HPC systems, especially the growing gap between CPU and GPU compute capabilities, as well as the introduction of low precision arithmetic in modern GPUs. We also describe MAGMA’s direct impact on several ECP projects. Maintaining portable performance across NVIDIA and AMD GPUs, and with current efforts toward supporting Intel GPUs, MAGMA ensures its adaptability and relevance in the ever-evolving landscape of GPU architectures.

97 MATHEMATICS AND COMPUTING↗

Prompt Phrase Ordering Using Large Language Models in HPC: Evaluating Prompt Sensitivity

Large language models (LLMs) have demonstrated effective performance in domain-specific tasks, often requiring a well-designed prompt to guide their responses. However, optimizing the right prompt is challenging due to prompt sensitivity—the phenomenon where small changes in the prompt can lead to significant variations in performance. In this study, we evaluate prompt performance by examining all permutations of independent phrases to investigate prompt sensitivity and robustness. We used two datasets: the GSM8k dataset, which assesses mathematical reasoning, and a custom template prompt for summarizing database metadata. Our goal was to evaluate the performance across all permutations of a sequence of prompt phrases. The study was conducted using the llama3-instruct- 7B model hosted on Ollama, with computations parallelized in a high-performance computing environment. By comparing the average index of phrases in the best and worst-performing prompts, we found that the order of independent phrases within a prompt significantly impacts LLM performance. Additionally, we used Hamming distance to assess changes between phrase orderings, concluding that prompt modifications can dramatically affect scores, often by almost random chance. These findings support existing research on prompt sensitivity. We discuss the challenges of prompt optimization, noting that altering phrases in a successful prompt does not always result in another successful prompt.

97 MATHEMATICS AND COMPUTING↗

Dynamic Distribution of High-Rate Data Processing from CERN to Remote HPC Data Centers

The prompt reconstruction of the data recorded from the Large Hadron Collider (LHC) detectors has always been addressed by dedicated resources at the CERN Tier-0. Such workloads come in spikes due to the nature of the operation of the accelerator and in special high load occasions experiments have commissioned methods to distribute (spill-over) a fraction of the load to sites outside CERN. The present work demonstrates a new way of supporting the Tier-0 environment by provisioning resources elastically for such spilled-over workflows onto the Piz Daint Supercomputer at CSCS. Furthermore, this is implemented using containers, tuning the existing batch scheduler and reinforcing the scratch file system, while still using standard Grid middleware. ATLAS, CMS and CSCS have jointly run selected prompt data reconstruction on up to several thousand cores on Piz Daint into a shared environment, thereby probing the viability of the CSCS high performance computer site as on demand extension of the CERN Tier-0, which could play a role in addressing the future LHC computing challenges for the high luminosity LHC.

97 MATHEMATICS AND COMPUTING↗

Hvac: Removing I/O Bottleneck for Large-Scale Deep Learning Applications

Scientific communities are increasingly adopting deep learning (DL) models in their applications to accelerate scientific discovery processes. However, with rapid growth in the computing capabilities of HPC supercomputers, large-scale DL applications have to spend a significant portion of training time performing I/O to a parallel storage system. Previous research works have investigated optimization techniques such as prefetching and caching. Unfortunately, there exist non-trivial challenges to adopting the existing solutions on HPC supercomputers for large-scale DL training applications, which include non-performance and/or failures at extreme scale, lack of portability and generality in design, complex deployment methodology, and being limited to a specific application or dataset. To address these challenges, we propose High-Velocity AI Cache (HVAC), a distributed read-cache layer that targets and fully exploits the node-local storage or near node-local storage technology. HVAC seamlessly accelerates read I/O by aggregating node-local or near node-local storage, avoiding metadata lookups and file locking while preserving portability in the application code. We deploy and evaluate HVAC on 1,024 nodes (with over 6000 NVIDIA V100 GPUS) of the Summit supercomputer. In particular, we evaluate the scalability, efficiency, accuracy, and load distribution of HVAC compared to GPFS and XFS-on-NVMe. With four different DL applications, we observe an average 25 % performance improvement atop GPFS and 9% drop against XFS-on-NVMe, which scale linearly and are considered the performance upper bound. We envision HVAC as an important caching library for upcoming HPC supercomputers such as Frontier.

Khan, Awais↗

LDMS-GPU: Lightweight Distributed Metric Service (LDMS) for NVIDIA GPGPUs

GPUs are now a fundamental accelerator for many high-performance computing applications. They are viewed by many as a technology facilitator for the surge in fields like machine learning and Convolutional Neural Networks. To deliver the best performance on a GPU, we need to create monitoring tools to ensure that we optimize the code to get the most performance and efficiency out of a GPU. Since NVIDIA GPUs are currently the most commonly implemented in HPC applications and systems, NVIDIA tools are the solution for performance monitoring. The Light-Weight Distributed Metric System (LDMS) at Sandia is an infrastructure widely adopted for large-scale systems and application monitoring. Sandia has developed CPU application monitoring capability within LDMS. Therefore, we chose to develop a GPU monitoring capability within the same framework. In this report, we discuss the current limitations in the NVIDIA monitoring tools, how we overcame such limitations, and present an overview of the tool we built to monitor GPU performance in LDMS and its capabilities. Also, we discuss our current validation results. Most of the performance counter results are the same in both vendor tools and our tool when using LDMS to collect these results. Furthermore, our tool provides these statistics during the entire runtime of the tool as a time series and not just aggregate statistics at the end of the application run. This allows the user to see the progress of the behavior of the applications during their lifetime.

97 MATHEMATICS AND COMPUTING↗