Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “compute in memory”

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 883 records · Page 49
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Optimizing High Performance Markov Clustering for Pre-Exascale Architectures

HipMCL is a high-performance distributed memory implementation of the popular Markov Cluster Algorithm (MCL) and can cluster large-scale networks within hours using a few thousand CPU-equipped nodes. It relies on sparse matrix computations and heavily makes use of the sparse matrix-sparse matrix multiplication kernel (SpGEMM). The existing parallel algorithms in HipMCL are not scalable to Exascale architectures, both due to their communication costs dominating the runtime at large concurrencies and also due to their inability to take advantage of accelerators that are increasingly popular. In this work, we systematically remove scalability and performance bottlenecks of HipMCL. We enable GPUs by performing the expensive expansion phase of the MCL algorithm on GPU. Additionally, we propose a CPU-GPU joint distributed SpGEMM algorithm called pipelined Sparse SUMMA and integrate a probabilistic memory requirement estimator that is fast and accurate. Furthermore, we develop a new merging algorithm for the incremental processing of partial results produced by the GPUs, which improves the overlap efficiency and the peak memory usage. We also integrate a recent and faster algorithm for performing SpGEMM on CPUs. We validate our new algorithms and optimizations with extensive evaluations. With the enabling of the GPUs and integration of new algorithms, HipMCL is up to 12.4x faster, being able to cluster a network with 70 million proteins and 68 billion connections just under 15 minutes using 1024 nodes of ORNL's Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Integrating PGAS and MPI-based Graph Analysis

This project demonstrates that Chapel programs can interface with MPI-based libraries written in C++ without storing multiple copies of shared data. Chapel is a language for productive parallel computing using global address spaces (PGAS). We identified two approaches to interface Chapel code with the MPI-based Grafiki and Trilinos libraries. The first uses a single Chapel executable to call a C function that interacts with the C++ libraries. The second uses the mmap function to allow separate executables to read and write to the same block of memory on a node. We also encapsulated the second approach in Docker/Singularity containers to maximize ease of use. Comparisons of the two approaches using shared and distributed memory installations of Chapel show that both approaches provide similar scalability and performance.

97 MATHEMATICS AND COMPUTING↗
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↗
DOE OSTI2022

ASAP: Automatic Synthesis of Area-Efficient and Precision-Aware CGRAs

Coarse-grained reconfigurable accelerators (CGRAs) are a promising accelerator design choice that strikes a balance between performance and adaptability to different computing patterns across various applications domains. Designing a CGRA for a specific application domain involves enormous software/hardware engineering effort. Recent research works explore loop transformations, functional unit types, network topology, and memory size to identify optimal CGRA designs given a set of kernels from a specific application do- main. Unfortunately, the impact of functional units with different precision support has rarely been investigated. To address this gap, we propose ASAP – a hardware/software co-design framework that automatically identifies and synthesizes optimal precision-aware CGRA for a set of applications of interest. Our evaluation shows that ASAP generates specialized designs 3.2×, 4.21×, and 5.8× more efficient (in terms of performance per unit of energy or area) than non-specialized homogeneous CGRAs, for the scientific computing, embedded, and edge machine learning domains, respectively, with limited accuracy loss. Moreover, ASAP provides more efficient designs than other state-of-the-art synthesis frameworks for specialized CGRAs.

artificial intelligence↗
DOE OSTI2021

Feature Analysis, Tracking, and Data Reduction: An Application to Multiphase Reactor Simulation MFiX-Exa for In-Situ Use Case

As we enter the exascale computing regime, powerful supercomputers continue to produce much higher amounts of data than what can be stored for offline data processing. To utilize such high compute capabilities on these machines, much of the data processing needs to happen in situ, when the full high-resolution data is available at the supercomputer memory. In this article, we discuss our MFiX-Exa simulation, which models multiphase flow by tracking a very large number of particles through the simulation domain. In one of the use cases, the carbon particles interact with air to produce carbon dioxide bubbles from the reactor. These bubbles are of primary interest to the domain experts for these simulations. For this particle-based simulation, we propose a streaming technique that can be deployed in situ to efficiently identify the bubbles, track them over time, and use them to down-sample the data with minimal loss in these features.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Monitor support on accelerated processing device

A technique for implementing synchronization monitors on an accelerated processing device (“APD”) is provided. Work on an APD includes workgroups that include one or more wavefronts. All wavefronts of a workgroup execute on a single compute unit. A monitor is a synchronization construct that allows workgroups to stall until a particular condition is met. Responsive to all wavefronts of a workgroup executing a wait instruction, the monitor coordinator records the workgroup in an “entry queue.” The workgroup begins saving its state to a general APD memory and, when such saving is complete, the monitor coordinator moves the workgroup to a “condition queue.” When the condition specified by the wait instruction is met, the monitor coordinator moves the workgroup to a “ready queue,” and, when sufficient resources are available on a compute unit, the APD schedules the ready workgroup for execution on a compute unit.

42 ENGINEERING↗
DOE OSTI2022

A Scalable Pipeline for Gigapixel Whole Slide Imaging Analysis on Leadership Class HPC Systems

Whole Slide Imaging (WSI) captures microscopic details of a patient's histopathological features at multiple res-olutions organized across different levels. Images produced by WSI are gigapixel-sized, and saving a single image in memory requires a few gigabytes which is scarce since a complicated model occupies tens of gigabytes. Performing a simple met-ric operation on these large images is also expensive. High-performance computing (HPC) can help us quickly analyze such large images using distributed training of complex deep learning models. One popular approach in analyzing these images is to divide a WSI image into smaller tiles (patches) and then train a simpler model with these reduced-sized but large numbers of patches. However, we need to solve three pre-processing challenges efficiently for pursuing this patch-based approach. 1) Creating small patches from a high-resolution image can result in a high number (hundreds of thousands per image) of patches. Storing and processing these images can be challenging due to a large number of I/O and arithmetic operations. To reduce I/Oand memory accesses, an optimal balance between the size and number of patches must exist to reduce I/O and memory accesses. 2) WSI images may have tiny annotated regions for cancer tissue and a significant portion with normal and fatty tissues; correct patch sampling should avoid dataset imbalance. 3) storing and retrieving many patches to and from disk storage might incur I/O latency while training a deep learning model. An efficient distributed data loader should reduce I/O latency during the training and inference steps. This paper explores these three challenges and provides empirical and algorithmic solutions deployed on the Summit supercomputer hosted at the Oak Ridge Leadership Computing Facility.

Dash, Sajal↗
DOE OSTI2022

Computational discovery of two-dimensional rare-earth iodides: promising ferrovalley materials for valleytronics

Two-dimensional Ferrovalley materials with intrinsic valley polarization are rare but highly promising for valley-based nonvolatile random access memory and valley filter devices. These ferromagnetic materials exhibit valleys at or near the Fermi level with intrinsic magnetism. The strong coupling between magnetism and spin–orbit coupling induces intrinsic valley polarization. Using Kinetically Limited Minimization, an unconstrained crystal structure prediction algorithm, and prototype sampling based on first-principles calculations, we have discovered new Ferrovalley materials, rare-earth iodides RI 2 , where R is a rare-earth element belonging to Sc, Y, or La-Lu, and I is Iodine. The rare-earth iodides are layered and demonstrate either 2H, 1T, or 1T d phase as the ground state in bulk, analogous to transition metal dichalcogenides (TMDCs). The calculated exfoliation energy of monolayers (MLs) is comparable to that of graphene and TMDCs, suggesting possible experimental synthesis. The MLs in the 2H phase exhibit ferromagnetism due to unpaired electrons in d and f orbitals. Throughout the rare-earth series, d bands have valley polarization at K and $\overline{K}$ points in the Brillouin zone in the vicinity of the Fermi level. Large intrinsic valley polarization in the range of 15–143 meV without external stimuli is observed in these Ferrovalley materials, which can be enhanced further by applying an in-plane bi-axial strain. These valleys can selectively be probed and manipulated for information storage and processing, potentially offering superior performance beyond conventional electronics and spintronics. Here we further show that the 2H ferromagnetic phase of RI 2 MLs possesses non-zero Berry curvature and exhibits anomalous valley Hall effect with considerable anomalous Hall conductivity. Our work will incite exploratory synthesis of the predicted Ferrovalley materials and their application in valleytronics and beyond.

2D materials↗
DOE OSTI2021

Axom

Axom is an open-source library of "building block" software components that provide core infrastructure capabilities for HPC applications. Such capabilities include: input file parsing and verification for various file formats, format specification for material shape input to multi-material simulations and tools for placing those shapes on meshes, parallel distributed coordination of diagnostic messages, geometric primitives, spatial queries and spatial search acceleration data structures, in memory key-value data store for managing simulation data and parallel file I/O. Axom components are used widely across LLNL Advanced Simulation and Computing program applications. It is also part of the LLNL institutionally-supported RADIUSS project, which promotes and funds its adoption by projects across LLNL.

Zagaris, George↗
DOE OSTI2025

Parallel Programming in MCNP6

Monte Carlo N-Particle (MCNP)1 is a general-purpose Monte Carlo particle transport code developed by Los Alamos National Laboratory (LANL). To efficiently handle long simulations, MCNP version 6 (MCNP6) supports parallel execution using two primary programming models: • Shared-memory task-based threading using OpenMP (Open Multi-Processing), and • Distributed-memory calculations using MPI (Message Passing Interface). The OpenMP and MPI programming models enable MCNP6 to scale from desktop systems to high-performance computing (HPC) clusters, allowing users to run MCNP in one of three parallel modes: • OpenMP-only, • MPI-only, and • Hybrid (MPI + OpenMP). The choice of parallelization mode depends on the underlying computer architecture and the characteristics of the simulation problem.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

When ancient numerical demons meet physics-informed machine learning: adjoint-based gradients for implicit differentiable modeling

Recent advances in differentiable modeling, a genre of physics-informed machine learning that trains neural networks (NNs) together with process-based equations, have shown promise in enhancing hydrological models' accuracy, interpretability, and knowledge-discovery potential. Current differentiable models are efficient for NN-based parameter regionalization, but the simple explicit numerical schemes paired with sequential calculations (operator splitting) can incur numerical errors whose impacts on models' representation power and learned parameters are not clear. Implicit schemes, however, cannot rely on automatic differentiation to calculate gradients due to potential issues of gradient vanishing and memory demand. Here we propose a “discretize-then-optimize” adjoint method to enable differentiable implicit numerical schemes for the first time for large-scale hydrological modeling. The adjoint model demonstrates comprehensively improved performance, with Kling–Gupta efficiency coefficients, peak-flow and low-flow metrics, and evapotranspiration that moderately surpass the already-competitive explicit model. Therefore, the previous sequential-calculation approach had a detrimental impact on the model's ability to represent hydrological dynamics. Furthermore, with a structural update that describes capillary rise, the adjoint model can better describe baseflow in arid regions and also produce low flows that outperform even pure machine learning methods such as long short-term memory networks. The adjoint model rectified some parameter distortions but did not alter spatial parameter distributions, demonstrating the robustness of regionalized parameterization. Despite higher computational expenses and modest improvements, the adjoint model's success removes the barrier for complex implicit schemes to enrich differentiable modeling in hydrology.

58 GEOSCIENCES↗
DOE OSTI2020

Memory access system

A method and system for accessing a memory for a data processing system. The method comprises sending a read request for a plurality of locations in the memory to read the plurality of locations in parallel based on an upper bound for reading the memory. The upper bound for a number of locations is based on a group of constraints for the memory. The method receives a summed value of a plurality of memory values in the plurality of locations in the memory.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Redirecting data to improve page locality in a scalable data fabric

A data processing system includes a host processor, a local memory coupled to the host processor, a plurality of remote memory media, and a scalable data fabric coupled to the host processor and to the plurality of remote memory media. The scalable data fabric includes a filter for storing information indicating a location of data that is stored by the data processing system. The host processor includes a hardware sequencer coupled to the filter for selectively moving data stored by the filter to the local memory.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Compressing data for storage in cache memories in a hierarchy of cache memories

An electronic device includes at least one compression-decompression functional block and a hierarchy of cache memories with a first cache memory and a second cache memory. The at least one compression-decompression functional block receives data in an uncompressed state, compresses the data using one of a first compression or a second compression, and, after compressing the data, provides the data to the first cache memory for storage therein. When the data is retrieved from the first cache memory to be stored in the second cache memory, when the data is compressed using the first compression, the compression-decompression functional block decompresses the data to reverse effects of the first compression on the data, thereby restoring the data to the uncompressed state and provides the data compressed using the second compression or in the uncompressed state to the second cache memory for storage therein.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Pre-conditioned BFGS-based uncertainty quantification in elastic full-waveform inversion

SUMMARY Full-waveform inversion has become an essential technique for mapping geophysical subsurface structures. However, proper uncertainty quantification is often lacking in current applications. In theory, uncertainty quantification is related to the inverse Hessian (or the posterior covariance matrix). Even for common geophysical inverse problems its calculation is beyond the computational and storage capacities of the largest high-performance computing systems. In this study, we amend the Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm to perform uncertainty quantification for large-scale applications. For seismic inverse problems, the limited-memory BFGS (L-BFGS) method prevails as the most efficient quasi-Newton method. We aim to augment it further to obtain an approximate inverse Hessian for uncertainty quantification in FWI. To facilitate retrieval of the inverse Hessian, we combine BFGS (essentially a full-history L-BFGS) with randomized singular value decomposition to determine a low-rank approximation of the inverse Hessian. Setting the rank number equal to the number of iterations makes this solution efficient and memory-affordable even for large-scale problems. Furthermore, based on the Gauss–Newton method, we formulate different initial, diagonal Hessian matrices as pre-conditioners for the inverse scheme and compare their performances in elastic FWI applications. We highlight our approach with the elastic Marmousi benchmark model, demonstrating the applicability of pre-conditioned BFGS for large-scale FWI and uncertainty quantification.

58 GEOSCIENCES↗