Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Access Patterns”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

DUNE Database Development

The DUNE experiment will produce vast amounts of metadata, which describe the data coming from the read-out of the primary DUNE detectors. Various databases will make up the overall DB architecture for this metadata. ProtoDUNE at CERN is the largest existing prototype for DUNE and serves as a testing ground for - among other things - possible database solutions for DUNE. The subset of all metadata that is accessed during offline data reconstruction and analysis is referred to as ‘conditions data’ and it is stored in a dedicated database. As offline data reconstruction and analysis will be deployed on HTC and HPC resources, conditions data is expected to be accessed at very high rates. It is therefore crucial to store it in a granularity that matches the expected access patterns allowing for extensive caching. This requires a good understanding of the sources and use cases of conditions data. This contribution will briefly summarize the database architecture deployed at ProtoDUNE and explain the various sources of conditions data. We will present how the conditions data is retrieved and streamed from the databases and how it is handled to match expected access patterns.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

A Vision for Coupling Operation of US Fusion Facilities with HPC Systems and the Implications for Workflows and Data Management

The operation of large US Department of Energy (DOE) research facilities, like the DIII-D National Fusion Facility, results in the collection of complex multi-dimensional scientific datasets, both experimental and model-generated. In the future, it is envisioned that integrated data analysis coupled with large-scale high performance computing (HPC) simulations will be used to improve experimental planning and operation. Practically, massive data sets from these simulations provide the physics basis for generation of both reduced semi-analytic and machine-learning-based models. Storage of both HPC simulation datasets (generated from US DOE leadership computing facilities) and experimental datasets presents significant challenges. In this paper, we present a vision for a DOE-wide data management workflow that integrates US DOE fusion facilities with leadership computing facilities. Data persistence and long-term availability beyond the length of allocated projects is essential, particularly for verification and recalibration of artificial intelligence and machine learning (AI/ML) models. Because these data sets are often generated and shared among hundreds of users across multiple leadership computing facility centers, they would benefit from cross-platform accessibility, persistent identifiers (e.g. DOI, or digital object identifier), and provenance tracking. Here, the ability to handle different data access patterns suggests that a combination of low cost, high latency (e.g. for storing ML training sets) and high cost, low latency systems (e.g. for real-time, integrated machine control feedback) may be needed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Scalability Testing Approach for Internet of Things for Manufacturing SQL and NoSQL Database Latency and Throughput

The proliferation of low-cost sensors and industrial data solutions has continued to push the frontier of manufacturing technology. Machine learning and other advanced statistical techniques stand to provide tremendous advantages in production capabilities, optimization, monitoring, and efficiency. The tremendous volume of data gathered continues to grow, and the methods for storing the data are critical underpinnings for advancing manufacturing technology. This work aims to investigate the ramifications and design tradeoffs within a decoupled architecture of two prominent database management systems (DBMS): sql and NoSQL. A representative comparison is carried out with Amazon Web Services (AWS) DynamoDB and AWS Aurora MySQL. The technologies and accompanying design constraints are investigated, and a side-by-side comparison is carried out through high-fidelity industrial data simulated load tests using metrics from a major US manufacturer. The results support the use of simulated client load testing for comparing the latency of database management systems as a system scales up from the prototype stage into production. As a result of complex query support, MySQL is favored for higher-order insights, while NoSQL can reduce system latency for known access patterns at the expense of integrated query flexibility. Here, by reviewing this work, a manufacturer can observe that the use of high-fidelity load testing can reveal tradeoffs in IoTfM write/ingestion performance in terms of latency that are not observable through prototype-scale testing of commercially available cloud DB solutions.

AWS↗
DOE OSTI2024

Refactoring the elastic–viscous–plastic solver from the sea ice model CICE v6.5.1 for improved performance

This study focuses on the performance of the elastic–viscous–plastic (EVP) dynamical solver within the sea ice model, CICE v6.5.1. The study has been conducted in two steps. First, the standard EVP solver was extracted from CICE for experiments with refactored versions, which are used for performance testing. Second, one refactored version was integrated and tested in the full CICE model to demonstrate that the new algorithms do not significantly impact the physical results. The study reveals two dominant bottlenecks, namely (1) the number of Message Parsing Interface (MPI) and Open Multi-Processing (OpenMP) synchronization points required for halo exchanges during each time step combined with the irregular domain of active sea ice points and (2) the lack of single-instruction, multiple-data (SIMD) code generation. The standard EVP solver has been refactored based on two generic patterns. The first pattern exposes how general finite differences on masked multi-dimensional arrays can be expressed in order to produce significantly better code generation by changing the memory access pattern from random access to direct access. The second pattern takes an alternative approach to handle static grid properties. The measured single-core performance improvement is more than a factor of 5 compared to the standard implementation. The refactored implementation of strong scales on the Intel® Xeon® Scalable Processors series node until the available bandwidth of the node is used. For the Intel® Xeon® CPU Max series, there is sufficient bandwidth to allow the strong scaling to continue for all the cores on the node, resulting in a single-node improvement factor of 35 over the standard implementation. This study also demonstrates improved performance on GPU processors.

58 GEOSCIENCES↗
DOE OSTI2025

Data Placement Optimization for ATLAS in a Multi-Tiered Storage System within a Data Center

Scientific experiments and computations, especially in High Energy Physics, are generating and accumulating data at an unprecedented rate. Effectively managing this vast volume of data while ensuring efficient data analysis poses a significant challenge for data centers, which must integrate various storage technologies. This paper proposes addressing this challenge by designing and developing a precise data popularity prediction model utilizing state-of-theart AI/ML techniques. This model is crafted from the analysis of ATLAS data and access patterns. It enables us to migrate infrequently accessed data to more economical storage media, such as tape drives, while storing frequently accessed data on faster yet costlier storage media like HDD or SSD. This strategic approach ensures data is placed optimally into the appropriate storage classes, thereby maximizing storage capacity while minimizing data access latency for end-users. Furthermore, the paper includes a performance evaluation of the prediction model using various key metrics such as F1 score, accuracy, precision and recall. Finally, we present a prototype use case, leveraging real-world file access data to assess the model’s impact on performance.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

I/O in Machine Learning Applications on HPC Systems: A 360-degree Survey

Growing interest in Artificial Intelligence (AI) has resulted in a surge in demand for faster methods of Machine Learning (ML) model training and inference. This demand for speed has prompted the use of high performance computing (HPC) systems that excel in managing distributed workloads. Because data is the main fuel for AI applications, the performance of the storage and I/O subsystem of HPC systems is critical. In the past, HPC applications accessed large portions of data written by simulations or experiments or ingested data for visualizations or analysis tasks. ML workloads perform small reads spread across a large number of random files. This shift of I/O access patterns poses several challenges to modern parallel storage systems. In this paper, we survey I/O in ML applications on HPC systems, and target literature within a 6-year time window from 2019 to 2024. We define the scope of the survey, provide an overview of the common phases of ML, review available profilers and benchmarks, examine the I/O patterns encountered during offline data preparation, training, and inference, and explore I/O optimizations utilized in modern ML frameworks and proposed in recent literature. Lastly, we seek to expose research gaps that could spawn further R&D.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Design and Performance of Kokkos Staging Space toward Scalable Resilient Application Couplings

With the growing number of applications designed for heterogeneous HPC devices, application programmers and users are finding it challenging to compose scalable workflows as ensembles of these applications, that are portable, performant and resilient. The Kokkos C++ library has been designed to simplify this cumbersome procedure by providing an intra-application uniform programming model and portable performance. However, assembling multiple Kokkos-enabled applications into a complex workflow is still a challenge. Although Kokkos enables a uniform programming model, the inter-application data exchange still remains a challenge from both performance and software development cost perspectives. In order to address this issue, we propose Kokkos data staging memory space, an extension of Kokkos' data abstraction (memory space) for heterogeneous computing systems. This new abstraction allows to express data on a virtual shared-space for multiple Kokkos applications, thus extending Kokkos to support inter-application data exchange to build an efficient application workflow. Additionally, we study the effectiveness of asynchronous data layout conversions for applications requiring different memory access patterns for the shared data. Our preliminary evaluation with a synthetic benchmark indicate the effectiveness of this conversion adapted to three different scenarios representing access frequency and use patterns of the shared data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

EAP Patterns

EAP Patterns: Memory access and iteration patterns from the EAP code base with the physics removed. This is intended to be a serial app representing memory access patterns. We will populate the data structures from EAP output files to provide representative patterns of face and cell loops within the EAP code base The arguments to the program are the EAP output file and the number of MPI processors to emulate. Currently there is no MPI in the application and the number of processors represents a means of emulating the halo (clone) cells around the domain of the given processor. An optional argument `processor_ID` can be provided to specify the processor to emulate. In the future we plan to include the ability to run multiple MPI ranks to more accurately represent the on-node demands on memory bandwidth.

Swaminarayan, Sriram↗
DOE OSTI2024

Darshan for HEP applications

Modern HEP workflows must manage increasingly large and complex data collections. HPC facilities may be employed to help meet these workflows’ growing data processing needs. However, a better understanding of the I/O patterns and underlying bottlenecks of these workflows is necessary to meet the performance expectations of HPC systems.Darshan is a lightweight I/O characterization tool that captures concise views of HPC application I/O behavior. It intercepts application I/O calls at runtime, records file access statistics for each process, and generates log files detailing application I/O access patterns.Typical HEP workflows include event generation, detector simulation, event reconstruction, and subsequent analysis stages. A study of the I/O behavior of the ATLAS simulation and filtering stage, and the CMS simulation workflow using Darshan is presented, including insights into the I/O operations and data access size.

Wang, Rui↗
DOE OSTI2022

Understanding and Leveraging the I/O Patterns of Emerging Machine Learning Analytics

The scientific community is currently experiencing unprecedented amounts of data generated by cutting-edge science facilities. Soon facilities will be producing up to 1 PB/s which will force scientist to use more autonomous techniques to learn from the data. The adoption of machine learning methods, like deep learning techniques, in large-scale workflows comes with a shift in the workflow’s computational and I/O patterns. These changes often include iterative processes and model architecture searches, in which datasets are analyzed multiple times in different formats with different model configurations in order to find accurate, reliable and efficient learning models. This shift in behavior brings changes in I/O patterns at the application level as well at the system level. These changes also bring new challenges for the HPC I/O teams, since these patterns contain more complex I/O workloads. In this paper we discuss the I/O patterns experienced by emerging analytical codes that rely on machine learning algorithms and highlight the challenges in designing efficient I/O transfers for such workflows. We comment on how to leverage the data access patterns in order to fetch in a more efficient way the required input data in the format and order given by the needs of the application and how to optimize the data path between collaborative processes. We will motivate our work and show performance gains with a study case of medical applications.

Gainaru, Ana↗
DOE OSTI2020

XPlacer/Tracer

XPlacer/Tracer is a dynamic analysis tool that finds bad memory access patterns in heterogeneous codes. XPlacer/Tracer consists of two components. (1) A configurable ROSE plugin for source code instrumentation of heterogeneous code written in the C/C++ and CUDA programming languages. (2) A sample configuration that instruments memory accesses to dynamic memory in CPU and GPU codes, and a runtime library that tracks these memory accesses at runtime. The collected information is reported in form of a textual summary or as memory map that can be converted to images.

Pirkelbauer, PeterM.↗
DOE OSTI2023

Effectiveness and predictability of in-network storage cache for Scientific Workflows

Large scientific collaborations often have multiple scientists accessing the same set of files while doing different analyses, which create repeated accesses to the large amounts of shared data located far away. These data accesses have long latency due to distance and occupy the limited bandwidth available over the wide-area network. To reduce the wide-area network traffic and the data access latency, regional data storage caches have been installed as a new networking service. To study the effectiveness of such a cache system in scientific applications, we examine the Southern California Petabyte Scale Cache for a high-energy physics experiment. By examining about 3TB of operational logs, we show that this cache removed 67.6% of file requests from the wide-area network and reduced the traffic volume on wide-area network by 12. 3TB (or 35.4%) an average day. The reduction in the traffic volume (35.4%) is less than the reduction in file counts (67.6%) because the larger files are less likely to be reused. Due to this difference in data access patterns, the cache system has implemented a policy to avoid evicting smaller files when processing larger files. We also build a machine learning model to study the predictability of the cache behavior. Tests show that this model is able to accurately predict the cache accesses, cache misses, and network throughput, making the model useful for future studies on resource provisioning and planning.

Sim, Caitlin↗
DOE OSTI2024

Distributed Multi-GPU Community Detection on Exascale Computing Platforms

Community detection is a fundamental operation in graph mining, and by uncovering hidden structures and patterns within complex systems it helps solve fundamental problems pertaining to social networks, such as information diffusion, epidemics, and recommender systems. Scaling graph algorithms for massive networks becomes challenging on modern distributed-memory multi-GPU (Graphics Processing Unit) systems due to limitations such as irregular memory access patterns, load imbalances, higher communication-computation ratios, and cross-platform support. We present a novel algorithm HiPDPL-GPU (distributed parallel Louvain) to address these challenges. We conduct experiments involving different partitioning techniques to achieve optimized performance of HiPDPL-GPU on the two largest supercomputers: Frontier and Summit. Remarkably, HiPDPL-GPU processes a graph with 4.2 billion edges in less than 3 minutes using 1024 GPUs. Qualitatively performance of HiPDPL-GPU is similar or better compared to other state-of-the-art CPU- and GPU-based implementations. While prior GPU implementations have predominantly employed CUDA, our first-of-its-kind implementation for community detection is cross-platform, accommodating both AMD and NVIDIA GPUs.

graph algorithms, high performance comptuing↗
DOE OSTI2024

Distributed Multi-GPU Community Detection on Exascale Computing Platforms

Community detection is a fundamental operation in graph mining, and by uncovering hidden structures and patterns within complex systems it helps solve fundamental problems pertaining to social networks, such as information diffusion, epidemics, and recommender systems. Scaling graph algorithms for massive networks becomes challenging on modern distributed-memory multi-GPU (Graphics Processing Unit) systems due to limitations such as irregular memory access patterns, load imbalances, higher communication-computation ratios, and cross-platform support. We present a novel algorithm HiPDPL-GPU (Distributed Parallel Louvain) to address these challenges. We conduct experiments involving different partitioning techniques to achieve an optimized performance of HiPDPL-GPU on the two largest supercomputers: Frontier and Summit. Remarkably, HiPDPL-GPU processes a graph with 4.2 billion edges in less than 3 minutes using 1024 GPUs. Qualitatively, the performance of HiPDPL-GPU is similar or better compared to other state-of-the-art CPU- and GPU-based implementations. While prior GPU implementations have predominantly employed CUDA, our first-of-its-kind implementation for community detection is cross-platform, accommodating both AMD and NVIDIA GPUs.

Sattar, Naw Safrin↗
DOE OSTI2023

Highly-scalable GPU-accelerated compressible reacting flow solver for modeling high-speed flows

Emerging supercomputing systems utilize a combination of central processing units (CPUs) and graphics processing units (GPUs) in an effort to reach exascale capabilities while minimizing the energy footprint of operating such systems. Such heterogeneous machines introduce new challenges for fluids solvers because the hardware architecture and operation of a GPU are fundamentally different from conventional CPUs. In this work, a general approach for efficient implementation of finite-volume based reacting flow solvers on such heterogeneous systems is presented. Three main challenges, namely, data access pattern, thread divergence, and thread safety, are addressed. Since compressible reacting flows require special methods to deal with chemical reactions, hyperbolic and nonlinear convection terms, and the presence of turbulence, specific algorithms that ensure GPU-based efficiency are developed. The approach is demonstrated on the widely available OpenFOAM open source software by modifying core algorithms for GPU accessibility. The scalability of the resulting solver, is demonstrated using practical test cases, including flow through a scramjet engine and the dynamics of a rotating detonation engine. Here, the solver provides near-ideal scaleup on a large number of GPUs (>3000), and extremely efficient use of the GPUs, with throughput nearly a constant even when processing a large number of control volumes.

42 ENGINEERING↗
DOE OSTI2022

Memory access optimization for particle operations in computational fluid dynamics-discrete element method simulations

Computational Fluid Dynamics - Discrete Element Method is used to model gas-solid systems in several applications in energy, pharmaceutical and petrochemical industries. Computational performance bottlenecks often limit the problem sizes that can be simulated at industrial scale. The data structures used to store several millions of particles in such large-scale simulations have a large memory footprint that does not fit into the processor cache hierarchies on current high-performance-computing platforms, leading to reduced computational performance. This paper specifically addresses this aspect of memory access bottlenecks in industrial scale simulations. The use of space-filling curves to improve memory access patterns is described and their impact on computational performance is quantified in both shared and distributed memory parallelization paradigms. The Morton space filling curve applied to uniform grids and k-dimensional tree partitions are used to reorder the particle data-structure thus improving spatial and temporal locality in memory. The performance impact of these techniques when applied to two benchmark problems, namely the homogeneous-cooling-system and a fluidized-bed, are presented. We report these optimization techniques lead to approximately two-fold performance improvement in particle focused operations such as neighbor-list creation and data-exchange, with ~ 1.5 times overall improvement in a fluidization simulation with 1.27 million particles.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Instruction Roofline: An insightful visual performance model for GPUs

The Roofline performance model provides an intuitive approach to identify performance bottlenecks and guide performance optimization. However, the classic FLOP-centric approach is inappropriate for the emerging applications that perform more integer operations than floating point operations. In this article, we reintroduce our Instruction Roofline Model on NVIDIA GPUs and expand our evaluation of it. The Instruction Roofline incorporates instructions and memory transactions across all memory hierarchies together, and provides more performance insights than the FLOP-oriented Roofline Model, that is, instruction throughput, stride memory access patterns, bank conflicts, and thread predication. We use our Instruction Roofline methodology to analyze eight proxy applications: HPGMG from AMReX, Matrix Transpose benchmarks, ADEPT from MetaHipMer's sequence alignment phase, EXTENSION from MetaHipMer's local assembly phase, CUSP, cuSPARSE, cudaTensorCoreGemm, and cuBLAS. We demonstrate the ability of our methodology to understand various aspects of performance and performance bottlenecks on NVIDIA GPUs and motivate code optimizations.

Ding, N↗
DOE OSTI2024

Boosting RDataFrame performance with transparent bulk event processing

RDataFrame is ROOT’s high-level interface for Python and C++ data analysis. Since it first became available, RDataFrame adoption has grown steadily and it is now poised to be a major component of analysis software pipelines for LHC Run 3 and beyond. Thanks to its design inspired by declarative programming principles, RDataFrame enables the development of highperformance, highly parallel analyses without requiring expert knowledge of multi-threading and I/O: user logic is expressed in terms of self-contained, small computation kernels tied together by a high-level API. This design completely decouples analysis logic from its actual execution, and opens several interesting avenues for workflow optimization. In particular, in this work we explore the benefits of moving internal data processing from an event-by-event to a bulkby-bulk loop. This refactoring dramatically reduces the framework’s runtime overheads; in collaboration with the I/O layer it improves data access patterns; it exposes information that optimizing compilers might use to auto-vectorize the invocation of user-defined computations; finally, while existing user-facing interfaces remain unaffected, it becomes possible to additionally offer interfaces that explicitly expose bulks of events, useful e.g. for the injection of GPU kernels into the analysis workflow. In order to inform similar future R&D, design challenges will be presented, as well as an investigation of the relevant timememory trade-off backed by novel performance benchmarks.

Guiraud, Enrico↗