Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “heterogeneous memory systems”

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 OSTI2026

Risk-Aware Measurement Synchronization and Recovery for DSSE With Heterogeneous Data Sources

Power distribution systems are increasingly integrating heterogeneous sensors with varying data reporting rates and types, which pose challenges to achieving observability at the desired temporal resolution of distribution system state estimation (DSSE). Multisensor failures caused by extreme events exacerbate these issues, introducing substantial uncertainties into DSSE. This article proposes a novel solution to these challenges by ensuring high-resolution system observability despite heterogeneous data sources and multisensor failures. First, a deep learning architecture combining long short-term memory (LSTM) and graph convolutional network (GCN) is employed to synchronize meters with different reporting rates, aiming to achieve system observability. A random-walk-model-based approach is introduced to generate pseudo-measurements while properly characterizing their uncertainties under multisensor failures. Finally, a disaster-risk-informed observability metric (RiOM) is defined to quantify the uncertainty associated with state estimation results. The proposed framework offers deeper insights into the system observability on the fly compared with conventional analysis. The effectiveness of the framework is demonstrated on an IEEE standard test case and a large-scale real-world distribution feeder in mid-Minnesota in the U.S.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training

Deep Learning Recommendation Models (DLRMs) are critical applications in various domains and have evolved as one of the single largest machine learning applications. Trillions of DLRM parameters exceed the on-chip memory capacity of GPUs. Large-scale multi-node systems are required for distributed DLRM inference and training, which suffer from the all-to-all communication bottleneck, mainly limiting the scalability of ever-growing DLRMs. In recent years, SmartNICs have evolved with coupled computation and communication capabilities providing opportunities for a powerful heterogeneous device in the system. However, there isn't such a distributed system that fully leverages the abundant smartNIC resources that resolve the scalability issue of DLRMs. In this work, we proposed a software-hardware co-design of a heterogeneous smartNIC system that resolves the communication bottleneck of distributed DLRMs, mitigates the memory bandwidth pressure, and improves computation efficiency. We provide a set of smartNIC designs of cache systems (including local cache and remote cache) and smartNIC computation kernels which reduce data movement, relieve memory lookup intensity, and improve the GPU's computation efficiency. In addition, we propose a graph algorithm that improves the data locality of queries within batches which optimizes the overall system performance with higher data reuse. Our evaluation shows that our system achieves 2.1x latency speedup for inference and 1.6x throughput speedup for training.

Guo, Anqi↗
DOE OSTI2023

An organic electrochemical transistor for multi-modal sensing, memory and processing

By integrating sensing, memory and processing functionalities, biological nervous systems are energy and area efficient. Emulating such capabilities in artificial systems is, however, challenging and is limited by the device heterogeneity of sensing and processing cores. Here we report an organic electrochemical transistor capable of sensing, memory and processing. The device has a vertical traverse architecture and a crystalline–amorphous channel that can be selectively doped by ions to enable two reconfigurable modes: a volatile receptor and a non-volatile synapse. As a volatile receptor, the device is capable of multi-modal sensing and is responsive to stimuli such as ions and light. As a non-volatile synapse, it is capable of 10-bit analogue states, low switching stochasticity and good state retention. We also show that the homogeneous integration of the devices could provide functions such as conditioned reflexes and could be used for real-time cardiac disease diagnoses via reservoir computing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Distributed out-of-memory NMF on CPU/GPU architectures

We propose an efficient distributed out-of-memory implementation of the non-negative matrix factorization (NMF) algorithm for heterogeneous high-performance-computing systems. The proposed implementation is based on prior work on NMFk, which can perform automatic model selection and extract latent variables and patterns from data. In this work, we extend NMFk by adding support for dense and sparse matrix operation on multi-node, multi-GPU systems. The resulting algorithm is optimized for out-of-memory problems where the memory required to factorize a given matrix is greater than the available GPU memory. Memory complexity is reduced by batching/tiling strategies, and sparse and dense matrix operations are significantly accelerated with GPU cores (or tensor cores when available). Input/output latency associated with batch copies between host and device is hidden using CUDA streams to overlap data transfers and compute asynchronously, and latency associated with collective communications (both intra-node and inter-node) is reduced using optimized NVIDIA Collective Communication Library (NCCL) based communicators. Benchmark results show significant improvement, from 32X to 76x speedup, with the new implementation using GPUs over the CPU-based NMFk. Good weak scaling was demonstrated on up to 4096 multi-GPU cluster nodes with approximately 25,000 GPUs when decomposing a dense 340 Terabyte-size matrix and an 11 Exabyte-size sparse matrix of density 10 -6 .

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Modeling Analog Tile-Based Accelerators Using SST

Analog computing has been widely proposed to improve the energy efficiency of multiple important workloads including neural network operations, and other linear algebra kernels. To properly evaluate analog computing and explore more complex workloads such as systems consisting of multiple analog data paths, system level simulations are required. Moreover, prior work on system architectures for analog computing often rely on custom simulators creating signficant additional design effort and complicating comparisons between different systems. To remedy these issues, this report describes the design and implementation of a flexible tile-based analog accelerator element for the Structural Simulation Toolkit (SST). The element focuses on heavily on the tile controller—an often neglected aspect of prior work—that is sufficiently versatile to simulate a wide range of different tile operations including neural network layers, signal processing kernels, and generic linear algebra operations without major constraints. The tile model also interoperates with existing SST memory and network models to reduce the overall development load and enable future simulation of heterogeneous systems with both conventional digital logic and analog compute tiles. Finally, both the tile and array models are designed to easily support future extensions as new analog operations and applications that can benefit from analog computing are developed.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Toward Performance Portable Programming for Heterogeneous System-on-Chips: Case Study with Qualcomm Snapdragon SoC

Future heterogeneous Domain-Specific System-on-Chips (DSSoC) will be extraordinarily complex in terms of processors, memory hierarchies, and interconnection networks.To manage this complexity, architects, system software designers, and application developers need programming technologies that are flexible, accurate, efficient, and productive. These technologies will need to be as independent of any one specific architecture as is practical, because the sheer dimensionality and scale of the complexity will not allow porting and optimizing applications foreach given DSSoC. To address these issues, we are developing Cosmic Castle, a performance portable programming toolchain for streaming applications on heterogeneous architectures. The primary focus of Cosmic Castle is on enabling efficient and performant code generation through the smart compiler and intelligent runtime system. This paper presents the preliminary evaluation of our ongoing work toward Cosmic Castle. Specifically, we detail our code porting efforts and evaluate various benchmarks on the Qualcomm Snapdragon SoC using tools developed through Cosmic Castle.

Cabrera, Anthony↗
DOE OSTI2024

Efficient Parallelization of Irregular Applications on GPU Architectures

With the enlarging computation capacity of general Graphics Processing Units (GPUs), leveraging GPUs to accelerate parallel applications has become a critical topic in academia and industry. However, a wide range of irregular applications with the computation-/memory-intensive nature cannot easily achieve high GPU utilization. The challenges mainly involve the following aspects: first, data dependence leads to coarse-grained kernel and inefficient parallelism; second, heavy GPU memory usage may cause frequent memory evictions and extra overhead of I/O; third, specific computation patterns produce memory redundancies; last, workload balance and data reusability conjunctly benefit the overall performance, but there may exist a dynamic trade-off between them. Targeting these challenges, this dissertation proposes multiple optimizations to accelerate two real-world applications: many-body correlation functions to simulate nuclear physics in a large-scale scientific system; the other is the eALS-based matrix factorization recommendation system. To accelerate the calculations of many-body correlation functions, this dissertation presents three frameworks in GPU memory management and multi-GPU scheduling. Firstly, an optimized systematic GPU memory management framework, MemHC, utilizes a series of new memory reduction designs in GPU memory allocation, CPU/GPU communications, and GPU memory oversubscription. Secondly, an enhanced multi-GPU scheduling framework, MICCO, particularly by taking both data dimension (e.g., data reuse and data eviction) and computation dimension into account. MICCO designs a heuristic scheduling algorithm and a machine learning-based regression model to generate the optimal settings of a proposed new concept to manage the trade-off. Thirdly, a locality-aware multi-GPU scheduling framework. This scheduler leverages pipeline batch generation with a looking-ahead strategy by building local dependency graphs for memory transfer reduction and better data reuse, achieving up to 79.92% memory cost reduction and 1.67x speedup. To parallelize the eALS-based recommendation system, this dissertation proposes an efficient CPU/GPU heterogeneous recommendation system, HEALS. HEALS employs newly designed architecture-adaptive data formats to achieve load balance and good data locality on CPU and GPU. To mitigate the data dependence, HEALS presents a CPU/GPU collaboration model for both task parallelism and data parallelism with multiple kernel computation optimizations. In summary, this dissertation efficiently accelerates two typical irregular applications on GPUs by building four frameworks, including CPU/GPU collaboration, GPU memory management, and multi-GPU scheduling.

Wang, Qihan↗
DOE OSTI2022

Design and analysis of CXL performance models for tightly-coupled heterogeneous computing

Truly heterogeneous systems enable partitioned workloads to be mapped to the hardware that nets the best performance. However, current practice requires that inter-device communication between different vendors' hardware use host memory as an intermediary step. To date, there are no widely adopted solutions that allow accelerators to directly transfer data. A new cache-coherent protocol, CXL, aims to facilitate easier, fine-grained sharing between accelerators. In this work we analyze existing methods for designing heterogeneous applications that target GPUs and FPGAs working collaboratively, followed by an exploration to show the benefits of a CXL-enabled system. Specifically, we develop a test application that utilizes both an NVIDIA P100 GPU and a Xilinx U250 FPGA to show current communication limitations. From this application, we capture overall execution time and throughput measurements on the FPGA and GPU. We use these measurements as inputs to novel CXL performance models to show that using CXL caching instead of host memory results in a 1.31X speedup, while a more tightly-coupled pipelined implementation using CXL-enabled hardware would result in a speedup of 1.45X.

Cabrera, Anthony↗
NASA NTRS1994

High Performance Programming Using Explicit Shared Memory Model on Cray T3D1

The Cray T3D system is the first-phase system in Cray Research, Inc.'s (CRI) three-phase massively parallel processing (MPP) program. This system features a heterogeneous architecture that closely couples DEC's Alpha microprocessors and CRI's parallel-vector technology, i.e., the Cray Y-MP and Cray C90. An overview of the Cray T3D hardware and available programming models is presented. Under Cray Research adaptive Fortran (CRAFT) model four programming methods (data parallel, work sharing, message-passing using PVM, and explicit shared memory model) are available to the users. However, at this time data parallel and work sharing programming models are not available to the user community. The differences between standard PVM and CRI's PVM are highlighted with performance measurements such as latencies and communication bandwidths. We have found that the performance of neither standard PVM nor CRI s PVM exploits the hardware capabilities of the T3D. The reasons for the bad performance of PVM as a native message-passing library are presented. This is illustrated by the performance of NAS Parallel Benchmarks (NPB) programmed in explicit shared memory model on Cray T3D. In general, the performance of standard PVM is about 4 to 5 times less than obtained by using explicit shared memory model. This degradation in performance is also seen on CM-5 where the performance of applications using native message-passing library CMMD on CM-5 is also about 4 to 5 times less than using data parallel methods. The issues involved (such as barriers, synchronization, invalidating data cache, aligning data cache etc.) while programming in explicit shared memory model are discussed. Comparative performance of NPB using explicit shared memory programming model on the Cray T3D and other highly parallel systems such as the TMC CM-5, Intel Paragon, Cray C90, IBM-SP1, etc. is presented.

Simon, Horst D.↗
DOE OSTI2021

The PetscSF Scalable Communication Layer

PetscSF, the communication component of the Portable, Extensible Toolkit for Scientific Computation (PETSc), is designed to provide PETSc's communication infrastructure suitable for exascale computers that utilize GPUs and other accelerators. PetscSF provides a simple application programming interface (API) for managing common communication patterns in scientific computations by using a star-forest graph representation. PetscSF supports several implementations based on MPI and NVSHMEM, whose selection is based on the characteristics of the application or the target architecture. An efficient and portable model for network and intra-node communication is essential for implementing large-scale applications. The Message Passing Interface, which has been the de facto standard for distributed memory systems, has developed into a large complex API that does not yet provide high performance on the emerging heterogeneous CPU-GPU-based exascale systems. Here, we discuss the design of PetscSF, how it can overcome some difficulties of working directly with MPI on GPUs, and we demonstrate its performance, scalability, and novel features.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

High Performance Adaptive Physics Refinement to Enable Large-Scale Tracking of Cancer Cell Trajectory

The ability to track simulated cancer cells through the circulatory system, important for developing a mechanistic understanding of metastatic spread, pushes the limits of today's supercomputers by requiring the simulation of large fluid volumes at cellular-scale resolution. To overcome this challenge, we introduce a new adaptive physics refinement (APR) method that captures cellular-scale interaction across large domains and leverages a hybrid CPU-GPU approach to maximize performance. Through algorithmic advances that integrate multi-physics and multi-resolution models, we establish a finely resolved window with explicitly modeled cells coupled to a coarsely resolved bulk fluid domain. In this work we present multiple validations of the APR framework by comparing against fully resolved fluid-structure interaction methods and employ techniques, such as latency hiding and maximizing memory bandwidth, to effectively utilize heterogeneous node architectures. Collectively, these computational developments and performance optimizations provide a robust and scalable framework to enable system-level simulations of cancer cell transport.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1995

Heterogeneous concurrent computing with exportable services

Heterogeneous concurrent computing, based on the traditional process-oriented model, is approaching its functionality and performance limits. An alternative paradigm, based on the concept of services, supporting data driven computation, and built on a lightweight process infrastructure, is proposed to enhance the functional capabilities and the operational efficiency of heterogeneous network-based concurrent computing. TPVM is an experimental prototype system supporting exportable services, thread-based computation, and remote memory operations that is built as an extension of and an enhancement to the PVM concurrent computing system. TPVM offers a significantly different computing paradigm for network-based computing, while maintaining a close resemblance to the conventional PVM model in the interest of compatibility and ease of transition Preliminary experiences have demonstrated that the TPVM framework presents a natural yet powerful concurrent programming interface, while being capable of delivering performance improvements of upto thirty percent.

Sunderam, Vaidy↗
DOE OSTI2020

Indicator-directed Dynamic Power Management for Iterative Workloads on GPU-Accelerated Systems

Modern high-performance and warehouse computing centers show strong interest in minimizing system power consumption while satisfying customers’ quality of service (QoS). Dynamic voltage and frequency scaling (DVFS) is effective for achieving this goal. Nevertheless, automating the process online and making it transparent to users must address three major challenges: (1) Complexity — today’s hardware components (e.g., CPUs, GPUs, memory, network, etc.) can be configured in several or dozens of frequency/voltage states for satisfying divergent system demands. Given their combination and the emergence of heterogeneity, searching the optimal configuration in the design space online can be timing consuming. (2) QoS guarantee — user-defined objectives such as power constraint and performance target must be monitored, predicted and ensured at the best effort. (3) Adaptability — various known and unknown workloads run on systems. Workloads characteristics should be quickly determined and configurations dynamically adjusted in accord with workloads and QoS. In this work, we focus on applications exhibiting an interesting feature – iterative or periodic, which is common among conventional HPC and emerging machine learning workloads. We propose an online dynamic power-performance (ODPP) management framework to dynamically adjust GPU DVFS configurations to meet performance and power objectives and constraints, without any code annotation or intrusion. Particularly, ODPP extracts the performance and power indicators for applications from their resources utilization profiles in a short episode. It further automatically constructs an accurate model that infers from the indicators how the application's performance and power vary with GPU core and memory frequencies. Aided with the model, for both seen and unseen applications, ODPP can quickly determine the most appropriate DVFS configuration for their execution. We evaluate ODPP on an NVIDIA GPU using multiple exascale computing (ECP) and deep learning applications.

Zou, Pengfei↗
DOE OSTI2023

symPACK: A GPU-Capable Fan-Out Sparse Cholesky Solver

Sparse symmetric positive definite systems of equations are ubiquitous in scientific workloads and applications. Parallel sparse Cholesky factorization is the method of choice for solving such linear systems. Therefore, the development of parallel sparse Cholesky codes that can efficiently run on today’s large-scale heterogeneous distributed-memory platforms is of vital importance. Modern supercomputers offer nodes that contain a mix of CPUs and GPUs. To fully utilize the computing power of these nodes, scientific codes must be adapted to offload expensive computations to GPUs. We present symPACK, a GPU-capable parallel sparse Cholesky solver that uses one-sided communication primitives and remote procedure calls provided by the UPC++ library. We also utilize the UPC++ "memory kinds" feature to enable efficient communication of GPU-resident data. We show that on a number of large problems, symPACK outperforms comparable state-of-the-art GPU-capable Cholesky factorization codes by up to 14x on the NERSC Perlmutter supercomputer.

Bellavita, Julian↗
DOE OSTI2023

A Task Based Approach for Co-Scheduling Ensemble Workloads on Heterogeneous Nodes

Scientific workflows consist of multiple, connected applications, with data and results flowing from one to another in a pipeline. Traditionally, such workflows are executed in sequential order, storing intermediate data in storage disks. Co-scheduling application workflows concurrently on the same compute nodes would greatly reduce the cost of moving data to/from storage and allow real-time analysis of intermediate results. Nevertheless, most parallel programming runtimes do not allow seamless integration of various applications in a scientific workflow, in part due to the complexity of managing data and resources. The situation is even more complicated for heterogeneous systems. In this work we extend the Minos Computing Library (MCL) runtime to accelerate pipe-lined and parallel workloads where multiple applications are running in the same system. MCL’s asynchronous task library and runtime dynamically manages resources to allow co-scheduling of multiple processes sharing heterogeneous resources. In addition, we design a custom ex- tension of the Open Compute Language (OpenCL) to enable multiple processes to share device memory. We enable MCL to coordinate these shared buffers to allow for easy, fast data sharing between applications. Using malleable micro-benchmarks and two application workflows that combine scientific simulation and AI-based analysis, we show that our method outperforms traditional approaches.

Index Terms—Parallel systems, Scheduling and Task ↗
DOE OSTI2021

Performance Improvements for the Griffin Transport Solvers

Griffin is a Multiphysics Object-Oriented Simulation Environment based reactor multiphysics analysis application jointly developed by Idaho National Laboratory and Argonne National Laboratory. Griffin includes a variety of deterministic radiation transport solvers for fixed source, k-eigenvalue, adjoint, and subcritical multiplication, as well as transient solvers for point-kinetics, improved quasi-static, and spatial dynamics. A code assessment performed in FY-20 identified two significant issues with the transport solvers in Griffin: first, the primary heterogeneous SN (discrete ordinates) transport solver based on continuous finite element methods required significant mesh refinement and higher memory usage compared to solvers based on the method of characteristic for equivalent accuracy. Second, the homogeneous PN (spherical harmonics expansion) transport solver did not adequately support polynomial refinement, which is a feature usually required for problems with spatial homogenization and pronounced streaming, typical in fast or gas-cooled reactor systems. To address the first issue, the development effort focused on the more promising discontinuous finite element method (DFEM)-based SN transport solver in Griffin. The addition of an asynchronous parallel transport sweeper and coarse mesh finite difference (CMFD) acceleration have rendered a superior heterogeneous SN transport capability for multiphysics problems that requires far less computing resources in terms of both CPU time and memory usage. This is demonstrated with typical thermal- and fast-spectrum reactor benchmark problems, including 2D Transient Reactor Test, 3D Advanced Burner Test Reactor (ABTR), and 2D and 3D Empire microreactor. For the second issue, the development effort focused on a new transport solver based on the hybrid finite element PN method (HFEM-PN), equivalent to the variational nodal method, as well as a new diffusion solver based on HFEM-Diffusion. This solver is intended for homogenized domains with multiphysics coupling (i.e., supports mesh displacement, seamless temperature feedback, etc.). Initial calculations with the HFEM-Diffusion implementation show very good parallel efficiency for the residual evaluations with the 2D ABTR benchmark. A future development effort will be centered on further improvements to the CMFD, HFEM-PN, and DFEM diffusion solvers to ensure Griffin meets performance and software quality assurance requirements for advanced reactor design and analysis.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
NASA NTRS1988

Programming model for distributed intelligent systems

A programming model and architecture which was developed for the design and implementation of complex, heterogeneous measurement and control systems is described. The Multigraph Architecture integrates artificial intelligence techniques with conventional software technologies, offers a unified framework for distributed and shared memory based parallel computational models and supports multiple programming paradigms. The system can be implemented on different hardware architectures and can be adapted to strongly different applications.

Sztipanovits, J.↗
DOE OSTI2021

pyDRESCALk

Modern data scientists are tasked to analyze ever-growing data sets with increasingly complex relationships. Tensor decompositions have come to play a central role in identifying underlying latent structures in higher-order data. The problem of fitting tensor models to different distributions is complicated by the combinations of size, dimensionality, and sparsity present in real world data. The situation demands efficient algorithms designed for shared-memory and distributed systems. This work will present new research that tackles these challenges on several different fronts, leveraging optimizations in numerical algorithms and sparse tensor representations in heterogeneous high performance computing environments.

Bhattarai, Manish↗