Search NASA⌕ Search

SEARCH · Search NASA

Results for “runtime performance”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

An Efficient Multicore Implementation of a Novel HSS-Structured Multifrontal Solver Using Randomized Sampling

Here, we present a sparse linear system solver that is based on a multifrontal variant of Gaussian elimination and exploits low-rank approximation of the resulting dense frontal matrices. We use hierarchically semiseparable (HSS) matrices, which have low-rank off-diagonal blocks, to approximate the frontal matrices. For HSS matrix construction, a randomized sampling algorithm is used together with interpolative decompositions. The combination of the randomized compression with a fast ULV HSS factoriz ation leads to a solver with lower computational complexity than the standard multifrontal method for many applications, resulting in speedups up to 7 fold for problems in our test suite. The implementation targets many-core systems by using task parallelism with dynamic runtime scheduling. Numerical experiments show performance improvements over state-of-the-art sparse direct solvers. The implementation achieves high performance and good scalability on a range of modern shared memory parallel systems, including the Intel Xeon Phi (MIC). The code is part of a software package called STRUMPACK - STRUctured Matrices PACKage, which also has a distributed memory component for dense rank-structured matrices.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

DIMPLES: Distributed Influence Maximization for Pandemic pLanning on Exascale Systems

We study exascale parallel algorithms for the selection of intervention or monitoring strategies in massive realistic socio-technical networks through scalable Influence Maximization (InfMax) algorithms. We employ novel techniques to enable efficient scaling on up to 8k nodes of OLCF Frontier, with 65k AMD GPUs and 458k AMD CPU cores. Current state-of-the-art InfMax tools are limited to networks with only a few million actors (vertices) and a few hundred million interactions (edges). By overcoming these limitations, we show that our approach is capable of processing a realistic social contact network of the United States with 285 million nodes and about 8 billion edges. This two orders-of-magnitude improvement over the previous state-of-the-art is obtained by leveraging algorithmic advancements for the InfMax problem and designing several problem-specific approaches to overlap communication with computation, improve GPU efficiency, and lower the application’s memory requirements. We evaluate strong scaling for computing 10k most influential seeds using up to 8k nodes of an exascale system, and weak scaling from 128 to 8k system nodes for seed sets ranging from 625 to 40k seeds. We achieve the fastest-known runtime of 25 minutes while performing 48 million diffusion simulations totaling 2.31 petabytes to identify 40k influential seeds using 8k nodes, and take 5.75 minutes to identify 10k seeds while using 4k nodes.

Minutoli, Marco [Pacific Northwest National Labora↗
DOE OSTI2015

Final report: Compiled MPI. Cost-Effective Exascale Application Development

This is the final report on Compiled MPI: Cost-Effective Exascale Application Development, and summarizes the results under this project. The project investigated runtime enviroments that improve the performance of MPI (Message-Passing Interface) programs; work at Illinois in the last period of this project looked at optimizing data access optimizations expressed with MPI datatypes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

A Methodology for the Rapid Development of Scalable HPC Data Services

There is growing evidence in the scientifc computing community that parallel file systems are not sufficient for all HPC storage workloads. This realization has motivated extensive research in new storage system designs. The question of which alternative we should turn towards implies that there could be a single answer satisfying a wide range of very diverse applications. We argue that such a generic solution does not exist. Instead, custom data services should be designed and tailored to the needs of specific applications on specific hardware. Furthermore, custom data services should be designed in close collaboration with users. In this paper, we present a methodology for the rapid development of such data services. This methodology promotes the design of reusable building blocks that can be composed together efficiently through a runtime based on high-performance threading, tasking, and remote procedure calls. We illustrate the success of our methodology by showcasing three examples of data services designed from the same building blocks, yet targeting entirely different applications.

Dorier, Mattieu↗
DOE OSTI2026

Classic and Quantum Task-Based Intelligent Runtime for QIRs Running on Multiple QPUs

High-performance computing systems are rapidly evolving into heterogeneous platforms that fuse quantum accelerators with traditional classical processing units (CPUs) and graphical processing units (GPUs). This convergence calls for runtimes capable of managing both classical and quantum workloads in a unified manner. We introduce an intelligent, task-based runtime that marries the Intelligent RuntIme System (IRIS) asynchronous scheduler with a quantum programming stack through the Quantum Intermediate Representation Execution Engine (QIR-EE). Our design allows programs written in the quantum intermediate representation (QIR) to be dispatched concurrently to a variety of back-ends, including multiple quantum simulators and nascent quantum processors, enabling genuine hybrid execution on a single node. To illustrate its practicality, we partition a 4-qubit and 20-qubit circuit into three sub-circuits using quantum circuit cutting via the QCut library. Each sub-circuit is simulated independently by the QIR-EE driver within IRIS, after which a classical post-processing step merges the simulation results to recover the outcome of the original full-circuit computation. This case study demonstrates how finer task granularity can enable the parallel execution and lower the simulation burden per quantum task while preserving overall accuracy, highlighting the feasibility of our hybrid approach.

Miniskar, Narasinga Rao [ORNL] (ORCID:000000018259↗
DOE OSTI2015

Acceleration of GPU-based Krylov solvers via data transfer reduction

Krylov subspace iterative solvers are often the method of choice when solving large sparse linear systems. At the same time, hardware accelerators such as graphics processing units continue to offer significant floating point performance gains for matrix and vector computations through easy-to-use libraries of computational kernels. However, as these libraries are usually composed of a well optimized but limited set of linear algebra operations, applications that use them often fail to reduce certain data communications, and hence fail to leverage the full potential of the accelerator. In this study, we target the acceleration of Krylov subspace iterative methods for graphics processing units, and in particular the Biconjugate Gradient Stabilized solver that significant improvement can be achieved by reformulating the method to reduce data-communications through application-specific kernels instead of using the generic BLAS kernels, e.g. as provided by NVIDIA’s cuBLAS library, and by designing a graphics processing unit specific sparse matrix-vector product kernel that is able to more efficiently use the graphics processing unit’s computing power. Furthermore, we derive a model estimating the performance improvement, and use experimental data to validate the expected runtime savings. Finally, considering that the derived implementation achieves significantly higher performance, we assert that similar optimizations addressing algorithm structure, as well as sparse matrix-vector, are crucial for the subsequent development of high-performance graphics processing units accelerated Krylov subspace iterative methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

IRIS-GNN: Leveraging Graph Neural Networks for Scheduling on Truly Heterogeneous Runtime Systems

The diversity of accelerators in computer systems poses significant challenges for software developers, such as managing vendor-specific compiler toolchains, code fragmentation requiring different kernel implementations, and performance portability issues. To address these, the Intelligent Runtime System (IRIS) was developed. IRIS works across various systems, from smartphones to supercomputers, enabling automatic performance scaling based on available accelerators. It introduces abstract tasks for seamless execution transitions between accelerators while ensuring memory consistency and task dependencies. Although IRIS simplifies system details, optimal dynamic scheduling still requires user input to understand workload structures. To address this, we introduce a new scheduling policy for IRIS, termed IRIS-GNN, which is the first IRIS hybrid policy that operates in conjunction with the dynamic policies. This policy employs a Graph-Neural Network (GNN) to conduct Graph Classification of any task graphs submitted to IRIS. This GNN analyzes the structure and attributes of the task graph, categorizing it as either locality, concurrency, or mixed. This classification subsequently guides the selection of the dynamic policy used by IRIS. We provide a comparison of the performance of IRIS-GNN against the complete spectrum of IRIS’s dynamic policies, assess the overhead introduced by the GNN within this scheduling framework, and ultimately explore its practical application in real-world scenarios.

Johnston, Beau↗
DOE OSTI2022

Performance assessment of ensembles of in situ workflows under resource constraints

Summary Scientific breakthroughs in biomolecular methods and improvements in hardware technology have shifted from a long‐running simulation to a large set of shorter simulations running simultaneously, called an ensemble. In an ensemble, simulations are usually coupled with analyses of data produced by the simulations. In situ methods can be used to analyze large volumes of data generated by scientific simulations at runtime (i.e., simulations and analyses are performed concurrently). In this work, we study the execution of ensemble‐based simulations paired with in situ analyses using in‐memory staging methods. Using an ensemble of molecular dynamics in situ workflows with multiple simulations and analyses, we first show that collecting traditional metrics such as makespan, instructions per cycle, memory usage, or cache miss ratio is not sufficient to characterize complex behaviors of ensembles. We propose a method to evaluate the performance of ensembles of workflows that captures multiple resource usage aspects: resource efficiency, resource allocation, and resource provisioning. Experimental results demonstrate that the proposed method can effectively distinguish the performance of different component placements in an ensemble with up to 32 ensemble members. By evaluating different co‐location scenarios, our proposed performance indicators demonstrate benefits of co‐locating simulation and coupled analyses within a compute node.

Do, Tu Mai Anh↗
DOE OSTI2018

Manage OpenMP GPU Data Environment Under Unified Address Space

OpenMP has supported the offload of computations to accelerators such as GPUs since version 4.0. A crucial aspect in OpenMP offloading is to manage the accelerator data environment. Currently, this has to be explicitly programmed by users, which is non-trival and often results in suboptimal performance. The unified memory feature available in recent GPU architectures introduces another option, implicit management. However, our experiments show that it incurs several performance issues, especially under GPU memory oversubscription. In this paper, we propose a compiler and runtime collaborative approach to manage OpenMP GPU data under unified memory. In our framework, the compiler performs data reuse analysis to assist runtime data management. The runtime combines static and dynamic information to make optimized data management decisions.We have implement the proposed technology in the LLVM framework. The evaluation shows our method can achieve significant performance improvement for OpenMP GPU offloading.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Domaine-Specific Runtime to Orchestrate Computation on Heterogeneous Platforms

Task-based runtime systems in the past have sought to exploit inherent asynchronicity in the application execution to reduce overall runtime. In the last decade focus shifted to supporting the heterogeneity that is increasingly prevalent in high-performance computing systems. However, existing task-based runtime systems, being general, come with a challenging set of issues such as the complexity of abstractions and overheads. And they still leave much of the burden of exposing the parallelism on the application developers who also have to fit their applications to the runtime systems’ interfaces. In this paper we take a different approach for applications to target heterogeneous systems through domain-specific run-times. Our design aspires to leverage the domain-specific knowledge of a focused class of scientific simulations to pragmatically orchestrate computations in the simulations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

ADVERT: An Asynchronous Runtime for Fine-Grained Network Systems

The Data Vortex Network, a novel fine-grained high-radix interconnect, provides the performance and injection rates required by the fine-grained, unpredictable communication patterns of irregular applications. This paper presents ADVERT, an asynchronous runtime system that provides performance and productivity for the fine-grained Data Vortex Network. ADVERT integrates a lightweight memory manager (DVMem) for the user accessible SRAM integrated in the network interface, and a communication library (DVComm) that implements active messaging primitives (remote execution, and, based on this, get and put routines). ADVERT hides the complexity of controlling all the network hardware features through the low-level Data Vortex programming interface, while providing comparable performance. We discuss ADVERT's design and present an analysis of three exemplar irregular kernels (Triangle Count, Page Rank, and GUPS). ADVERT can provide the communication substrate required to build emerging higher level asynchronous many tasking runtimes and partitioned global address space libraries on top of the Data Vortex Network.

Friese, Ryan D.↗
DOE OSTI2021

GANpiler

Focal Area(s): Rather than augment or replace physical models with machine learning, we instead preserve the existing models and augment the underlying code with faster surrogate models created by generative adversarial networks (GAN). To leverage the performance optimization, we also propose a runtime system and user interfaces that allow prediction and tracking of accumulated error as well as dynamic, per-process decision making as to which model (if any) to use for each iteration. Science Challenge: This proposal sits at the nexus of two hard problems. First, climate models based on machine learning will be, by their nature, difficult to trust once their predictions begin diverging from the consensus. Second, compilers and hardware have been making only incremental performance gains for decades. GPGPUs have provided a welcome performance boost for codes that can take advantage of them, but there is no similar technology on the horizon to provide the next performance leap.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2018

Argobots: A Lightweight Low-Level Threading and Tasking Framework

In the past few decades, a number of user-level threading and tasking models have been proposed in the literature to address the shortcomings of OS-level threads, primarily with respect to cost and flexibility. Current state-of-the-art user-level threading and tasking models, however, are either too specific to applications or architectures or are not as powerful or flexible. In this paper, we present Argobots, a lightweight, low-level threading and tasking framework that is designed as a portable and performant substrate for high-level programming models or runtime systems. Argobots offers a carefully designed execution model that balances generality of functionality with providing a rich set of controls to allow specialization by the user or high-level programming model. We describe the design, implementation, and optimization of Argobots and present integrations with three example high-level models: OpenMP, MPI, and co-located I/O service. Evaluations show that (1) Argobots outperforms existing generic threading runtimes; (2) our OpenMP runtime offers more efficient interoperability capabilities than production OpenMP runtimes do; (3) when MPI interoperates with Argobots instead of Pthreads, it enjoys reduced synchronization costs and better latency hiding capabilities; and (4) I/O service with Argobots reduces interference with co-located applications, achieving performance competitive with that of the Pthreads version.

Seo, Sangmin (ORCID:0000000295159836)↗
DOE OSTI2025

Structure‐Aware Representation Learning for Effective Performance Prediction

ABSTRACT Application performance is a function of several unknowns stemming from the interactions between the application, runtime, OS, and underlying hardware, making it challenging to model performance using deep learning techniques, especially without a large labeled dataset. Collecting such labeled longitudinal datasets can take weeks. Intuitively, developers could save analysis time during code development by taking a comparative approach between multiple applications. However, the unknown dynamic interactions between applications and execution environments make it difficult for deep learning‐based models to predict the performance of new applications. In this paper, we address these problems by presenting a labeled dataset for the community and taking a comparative analysis approach to explore the source code differences between different correct implementations of the same problem. This paper assesses the feasibility of using purely static information, for example, Abstract Syntax Tree (AST), of applications to predict performance change based on code structure. We evaluate several deep learning‐based representation learning techniques for source code and propose an architecture for the tree‐based Long Short‐Term Memory (LSTM) models to discover latent representations for a source code's hierarchical structure. We demonstrate that our proposed architecture enables feed‐forward predictive models to predict change in performance using source code with up to 84% accuracy.

Ramadan, Tarek [Department of Computer Science Tex↗
DOE OSTI2017

XPRESS: eXascale PRogramming Environment and System Software, Final Report

The University of Oregon (UO) was a partner in the XPRESS project together with Sandia National Laboratories (SNL), Indiana University (IU), Lawrence Berkeley National Laboratory (LBL), Louisiana State University (LSU), Oak Ridge National Laboratory (ORNL), University of Houston (UH), and University of North Carolina (RENCI). UO's objective in the project was to develop "APEX (Asynchronous Performance Environment for eXascale)": performance measurement, analysis, feedback and control infrastructure to asynchronous tasking runtimes, specifically HPX (LSU) and HPX-5 (IU). During the project period, our efforts focused on the need to measure complex tasking models where the traditional instrumentation methods would not work due to an absent/meaningless call-path context and runtime control of task pre-emption and resumption, usually on a different operating system thread. In addition to the challenges of performance measurement, we developed a framework to provide information from all levels of the stack (hardware, operating system, runtime, application) to a feedback & control mechanism for controlling concurrency, task granularity, networking parameters, and other aspects of execution in order to improve execution performance, throughput and/or efficiency. UO demonstrated these developments on leading edge HPC machines at large scale, including a 2017 Gordon Bell submission in which APEX was used to optimize key bottlenecks in an astrophysics application. The following final report summarizes our accomplishments during the XPRESS project.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2021

TRITON: A Multi-GPU open source 2D hydrodynamic flood model

A new open source multi-GPU 2D flood model called TRITON is presented in this work. The model solves the 2D shallow water equations with source terms using a time-explicit first order upwind scheme based on an Augmented Roe's solver that incorporates a careful estimation of bed strengths and a local implicit formulation of friction terms. Here, the scheme is demonstrated to be first order accurate, robust and able to solve for flows under various conditions. TRITON is implemented such that the model effectively utilizes heterogeneous architectures, from single to multiple CPUs and GPUs. Different test cases are shown to illustrate the capabilities and performance of the model, showing promising runtimes for large spatial and temporal scales when leveraging the computer power of GPUs. Under this hardware configuration, communication and input/output subroutines may impact the scalability. The code is developed under an open source license and can be freely downloaded in https://code.ornl.gov/hydro/triton.

2D flood model↗
DOE OSTI2018

Prescriptive provenance for streaming analysis of workflows at scale

Abstract- We extend our approach capturing and relating the provenance and performance metrics of computational workflows as a diagnostic tool for runtime optimization and placement. One important challenge is the volume of extracted data, both for performance metrics and provenance, even when specifying filters and focusing on quantities of interest in a simulation. We reduce this data by performing anomaly detection on streaming data and store provenance for the detected anomalies, an approach we call prescriptive provenance. This paper discusses the Chimbuko architecture enabling the approach. We present the use of a protein structure propagation workflow based on NWChemEx. We are testing algorithms for anomaly detection and present preliminary results here obtained with Local Outlier Factor. While scaling remains a challenge, these results show that our robust Chimbuko architecture for streaming analysis with prescriptive provenance is a promising approach.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Exploring the feasibility of lossy compression for PDE simulations

Checkpoint restart plays an important role in high-performance computing (HPC) applications, allowing simulation runtime to extend beyond a single job allocation and facilitating recovery from hardware failure. Yet, as machines grow in size and in complexity, traditional approaches to checkpoint restart are becoming prohibitive. Current methods store a subset of the application’s state and exploit the memory hierarchy in the machine. However, as the energy cost of data movement continues to dominate, further reductions in checkpoint size are needed. Lossy compression, which can significantly reduce checkpoint sizes, offers a potential to reduce computational cost in checkpoint restart. This article investigates the use of numerical properties of partial differential equation (PDE) simulations, such as bounds on the truncation error, to evaluate the feasibility of using lossy compression in checkpointing PDE simulations. Restart from a checkpoint with lossy compression is considered for a fail-stop error in two time-dependent HPC application codes: PlasComCM and Nek5000. Results show that error in application variables due to a restart from a lossy compressed checkpoint can be masked by the numerical error in the discretization, leading to increased efficiency in checkpoint restart without influencing overall accuracy in the simulation.

97 MATHEMATICS AND COMPUTING↗