Search NASA⌕ Search

SEARCH · Search NASA

Results for “Offloading”

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 217 records · Page 12

Targeting GPUs with OpenMP directives on Summit: A simple and effective Fortran experience

We use OpenMP to target hardware accelerators (GPUs) on Summit, a newly deployed supercomputer at the Oak Ridge Leadership Computing Facility (OLCF), demonstrating simplified access to GPU devices for users of our astrophysics code GenASiS and useful speedup on a sample fluid dynamics problem. We modify our workhorse class for data storage to include members and methods that significantly streamline the persistent allocation of and association to GPU memory. Users offload computational kernels with OpenMP target directives that are rather similar to constructs already familiar from multi-core parallelization. In this initial example we ask, “With a given number of Summit nodes, how fast can we compute with and without GPUs?”, and find total wall time speedups of ~ 12X. We also find reasonable weak scaling up to 8000 GPUs (1334 Summit nodes). We make available the source code from this work at https://github.com/GenASiS/GenASiS_Basics.

97 MATHEMATICS AND COMPUTING↗

Implementation of High-Order Multireference Coupled-Cluster Methods on Intel Many Integrated Core Architecture

Here we discuss the implementation of multireference coupled-cluster formalism with singles, doubles, and noniterative triples (MRCCSD(T)), which is capable of taking advantage of the processing power of the Intel Xeon Phi coprocessor. We discuss the integration of two levels of parallelism underlying the MRCCSD(T) implementation with computational kernels designed to offload the computationally intensive parts of the MRCCSD(T) formalism to Intel Xeon Phi coprocessors. Special attention is given to the enhancement of the parallel performance by task reordering that has improved load balancing in the noniterative part of the MRCCSD(T) calculations. We also discuss aspects regarding efficient optimization and vectorization strategies.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Field Programmable Gate Arrays for Enhancing the Speed and Energy Efficiency of Quantum Dynamics Simulations

We present the first application of field programmable gate arrays (FPGAs) as new, customizable hardware architectures for carrying out fast and energy-efficient quantum dynamics simulations of large chemical/material systems. Instead of tailoring the software to fixed hardware, which is the typical case for writing quantum chemistry code for central processing units (CPUs) and graphics processing units (GPUs), FPGAs allow us to directly customize the underlying hardware (even at the level of specific electrical signals in the circuit) to give a truly optimized computational performance for quantum dynamics calculations. By offloading the most intensive and repetitive calculations onto an FPGA, we show that the computational performance of our real-time electron dynamics calculations can even exceed that of optimized commercial mathematical libraries running on high-performance GPUs. In addition to this impressive computational speedup, we show that FPGAs are immensely energy-efficient and consume 4 times less energy than modern GPU or CPU architectures. These energy savings are a practical and important metric for supercomputing centers (many of which exceed over $1 million in power costs alone), as exascale computing capabilities become more widespread and commonplace. Taken together, the implementation techniques and performance metrics of our study demonstrate that FPGAs could play a promising role in upcoming quantum chemistry and materials science applications, particularly for the acceleration and energy-efficient execution of quantum dynamics calculations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Noncontact excitation of multi-GHz lithium niobate electromechanical resonators

Abstract The demand for high-performance electromechanical resonators is ever-growing across diverse applications, ranging from sensing and time-keeping to advanced communication devices. Among the electromechanical materials being explored, thin-film lithium niobate stands out due to its strong piezoelectric properties and low acoustic loss. However, in nearly all existing lithium niobate electromechanical devices, the configuration is such that the electrodes are in direct contact with the mechanical resonator. This configuration introduces an undesirable mass-loading effect, producing spurious modes and additional damping. Here, we present an electromechanical platform that mitigates this challenge by leveraging a flip-chip bonding technique to separate the electrodes from the mechanical resonator. By offloading the electrodes from the resonator, our approach yields a substantial increase in the quality factor of these resonators, paving the way for enhanced performance and reliability for their device applications.

Instruments & Instrumentation↗

Multichannel meta-imagers for accelerating machine vision

Rapid developments in machine vision technology have impacted a variety of applications, such as medical devices and autonomous driving systems. These achievements, however, typically necessitate digital neural networks with the downside of heavy computational requirements and consequent high energy consumption. As a result, real-time decision-making is hindered when computational resources are not readily accessible. Here we report a meta-imager designed to work together with a digital back end to offload computationally expensive convolution operations into high-speed, low-power optics. Further, in this architecture, metasurfaces enable both angle and polarization multiplexing to create multiple information channels that perform positively and negatively valued convolution operations in a single shot. We use our meta-imager for object classification, achieving 98.6% accuracy in handwritten digits and 88.8% accuracy in fashion images. Owing to its compactness, high speed and low power consumption, our approach could find a wide range of applications in artificial intelligence and machine vision applications.

47 OTHER INSTRUMENTATION↗

Parallel IO Libraries for Managing HEP Experimental Data

The computing and storage requirements of the energy and intensity frontiers will grow significantly during the Run 4 & 5 and the HL-LHC era. Similarly, in the intensity frontier, with larger trig ger readouts during supernovae explosions, the Deep Underground Neutrino Experiment (DUNE) will have unique computing challenges that could be addressed by the use of parallel and accelerated dataprocessing capabilities. Most of the requirements of the energy and intensity frontier experiments rely on increasing the role of high performance computing (HPC) in the HEP community. In this presentation, we will describe our ongoing efforts that are focused on using HPC resources for the next generation HEP experiments. The HEPCCE (High Energy Physics-Center for Computational Excellence) IOS (Input/Output and Storage) group has been developing approaches to map HEP data to the HDF5 , an IO library optimized for the HPC platforms to store the intermediate HEP data. The complex HEP data products are serialized using ROOT to allow for experiment independent general mapping approaches of the HEP data to the HDF5 format. The mapping approaches can be optimized for high performance parallel IO. Similarly, simpler data can be directly mapped into the HDF5, which can also be suitable for offloading into the GPUs directly. We will present our works on both complex and simple data model models.

Bashyal, Amit↗

Porting ATLAS Fast Calorimeter Simulation to GPUs with Performance Portable Programming Models

FastCaloSim is a parameterized simulation of the particle energy response and of the energy distribution in the ATLAS calorimeter. It is a relatively small and self-contained package with massive inherent parallelism and captures the essence of GPU offloading via important operations like data transfer, memory initialization, floating point operations, and reduction. It was identified by the High Energy Physics Center for Computational Excellence project as a good testbed for evaluating the performance and ease of portability of programming models. In this paper, we will discuss the results of our evaluation of the porting process to Kokkos, SYCL, Alpaka, OpenMP and std::par (nvc++), and compare performance on NVIDIA, AMD and Intel GPUs, as well as multicore CPUs.

97 MATHEMATICS AND COMPUTING↗

Accelerating detector simulations with Celeritas: Profiling and performance optimizations

Celeritas is a GPU-optimized Monte Carlo (MC) particle transport code designed to meet the growing computational demands of next-generation high energy physics (HEP) experiments. It provides efficient simulation of electromagnetic (EM) physics processes in complex geometries with magnetic fields, detector hit scoring, and seamless integration into Geant4-driven applications to offload EM physics to GPUs. Recent efforts have focused on performance optimizations and expanding profiling capabilities. This paper presents some key advancements, including the integration of the Perfetto system profiling tool for detailed performance analysis and the development of track-sorting methods to improve computational efficiency.

Lund, Amanda [Argonne National Laboratory (ANL)]↗

Excited-state electronic structure of molecules using many-body Green’s functions: Quasiparticles and electron–hole excitations with VOTCA-XTP

Here, we present the open-source VOTCA-XTP software for the calculation of the excited-state electronic structure of molecules using many-body Green’s function theory in the GW approximation with the Bethe–Salpeter equation (BSE). This work provides a summary of the underlying theory and discusses the details of its implementation based on Gaussian orbitals, including resolution-of-identity techniques and different approaches to the frequency integration of the self-energy or acceleration by offloading compute-intensive matrix operations using graphics processing units in a hybrid OpenMP/Cuda scheme. A distinctive feature of VOTCA–XTP is the capability to couple the calculation of electronic excitations to a classical polarizable environment on an atomistic level in a coupled quantum- and molecular-mechanics (QM/MM) scheme, where a complex morphology can be imported from Molecular Dynamics simulations. The capabilities and limitations of the GW–BSE implementation are illustrated with two examples. First, we study the dependence of optically active electron–hole excitations in a series of diketopyrrolopyrrole-based oligomers on molecular-architecture modifications and the number of repeat units. Second, we use the GW–BSE/MM setup to investigate the effect of polarization on localized and intermolecular charge-transfer excited states in morphologies of low-donor content rubrene–fullerene mixtures. These showcases demonstrate that our implementation currently allows us to treat systems with up to 2500 basis functions on regular shared-memory workstations, providing accurate descriptions of quasiparticle and coupled electron–hole excited states of various characters on an equal footing.

74 ATOMIC AND MOLECULAR PHYSICS↗

ddcMD: A fully GPU-accelerated molecular dynamics program for the Martini force field

We have implemented the Martini force field within Lawrence Livermore National Laboratory’s molecular dynamics program, ddcMD. The program is extended to a heterogeneous programming model so that it can exploit graphics processing unit (GPU) accelerators. In addition to the Martini force field being ported to the GPU, the entire integration step, including thermostat, barostat, and constraint solver, is ported as well, which speeds up the simulations to 278-fold using one GPU vs one central processing unit (CPU) core. A benchmark study is performed with several test cases, comparing ddcMD and GROMACS Martini simulations. The average performance of ddcMD for a protein–lipid simulation system of 136k particles achieves 1.04 µs/day on one NVIDIA V100 GPU and aggregates 6.19 µs/day on one Summit node with six GPUs. The GPU implementation in ddcMD offloads all computations to the GPU and only requires one CPU core per simulation to manage the inputs and outputs, freeing up remaining CPU resources on the compute node for alternative tasks often required in complex simulation campaigns.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

GPU acceleration of hybrid functional calculations in the SPARC electronic structure code

We present a Graphics Processing Unit (GPU)-accelerated version of the real-space SPARC electronic structure code for performing hybrid functional calculations in generalized Kohn–Sham density functional theory. In particular, we develop a batch variant of the recently formulated Kronecker product-based linear solver for the simultaneous solution of multiple linear systems. We then develop a modular, math kernel based implementation for hybrid functionals on NVIDIA architectures, where computationally intensive operations are offloaded to the GPUs, while the remaining workload is handled by the central processing units (CPUs). Considering bulk and slab examples, we demonstrate that GPUs enable up to 8× speedup in node-hours and 80× in core-hours compared to CPU-only execution, reducing the time to solution on V100 GPUs to around 300 s for a metallic system with over 6000 electrons, and significantly reducing the computational resources required for a given wall time.

Kohn-Sham density functional theory↗

Performance portability for the CMS Reconstruction with Alpaka

For CMS, Heterogeneous Computing is a powerful tool to face the computational challenges posed by the upgrades of the LHC, and will be used in production at the High Level Trigger during Run 3. In principle, to offload the computational work on non-CPU resources, while retaining their performance, different implementations of the same code are required. This would introduce code-duplication which is not sustainable in terms of maintainability and testability of the software. Performance portability libraries allow to write code once and run it on different architectures with close-to-native performance. The CMS experiment is evaluating performance portability libraries for the near term future.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Supporting multiple hardware architectures at CMS: the integration and validation of POWER9

Computing resources in the Worldwide LHC Computing Grid (WLCG) have been based entirely on the x86 architecture for more than two decades. In the near future, however, heterogeneous non-x86 resources, such as ARM, POWER and Risc-V, will become a substantial fraction of the resources that will be provided to the LHC experiments, due to their presence in existing and planned world-class HPC installations. The CMS experiment, one of the four large detectors at the LHC, has started to prepare for this situation, with the CMS software stack (CMSSW) already compiled for multiple architectures. In order to allow for a production use, the tools for workload management and job distribution need to be extended to be able to exploit heterogeneous architectures. Profiting from the opportunity to exploit the first sizable IBM Power9 allocation available on Marconi100 HPC system at CINECA, CMS developed all the needed modifications to the CMS workload management system. After a successful proof of concept, a full physics validation has been performed in order to bring the system in production. The experiences are of very high value, when it comes to commissioning of the similar (even larger) Summit HPC system at Oak Ridge, where CMS is also expecting a resource allocation. Moreover the compute power of those systems is being provided also via GPUs and this represents an extremely valuable opportunity to exploit the offloading capability already implemented in CMSSW. The status of the current integration including the exploitation of the GPUs, the results of the validation as well as the future plans will be shown and discussed.

Boccali, Tommaso [INFN, Pisa]↗

Web Proxy Auto Discovery for the WLCG

All four of the LHC experiments depend on web proxies (that is, squids) at each grid site to support software distribution by the CernVM FileSystem (CVMFS). CMS and ATLAS also use web proxies for conditions data distributed through the Frontier Distributed Database caching system. ATLAS & CMS each have their own methods for their grid jobs to find out which web proxies to use for Frontier at each site, and CVMFS has a third method. Those diverse methods limit usability and flexibility, particularly for opportunistic use cases, where an experiment’s jobs are run at sites that do not primarily support that experiment. This paper describes a new Worldwide LHC Computing Grid (WLCG) system for discovering the addresses of web proxies. The system is based on an internet standard called Web Proxy Auto Discovery (WPAD). WPAD is in turn based on another standard called Proxy Auto Configuration (PAC). Both the Frontier and CVMFS clients support this standard. The input into the WLCG system comes from squids registered in the ATLAS Grid Information System (AGIS) and CMS SITECONF files, cross-checked with squids registered by sites in the Grid Configuration Database (GOCDB) and the OSG Information Management (OIM) system, and combined with some exceptions manually configured by people from ATLAS and CMS who operate WLCG Squid monitoring. WPAD servers at CERN respond to http requests from grid nodes all over the world with a PAC file that lists available web proxies, based on IP addresses matched from a database that contains the IP address ranges registered to organizations. Large grid sites are encouraged to supply their own WPAD web servers for more flexibility, to avoid being affected by short term long distance network outages, and to offload the WLCG WPAD servers at CERN. The CERN WPAD servers additionally support requests from jobs running at non-grid sites (particularly for LHC@Home) which it directs to the nearest publicly accessible web proxy servers. Furthermore, the responses to those requests are geographically ordered based on a separate database that maps IP addresses to longitude and latitude.

97 MATHEMATICS AND COMPUTING↗

Online helium inventory monitoring of JLab cryogenic systems

There are five cryogenic plants at Jefferson Lab which support the LINAC, experiment hall end-stations and test facility. The majority of JLab's helium inventory, which is around 15 tons, is allocated in the LINAC cryo-modules, with the majority of the balance of helium distributed at the cryogenic-plant level mainly as stored gas and liquid for stable operation. Due to the organic evolution of the five plants and independent actions within the experiment halls, the traditional inventory management strategy suffers from rapid identification of potential leaks. This can easily result in losses many times higher than the normally accepted (average) loss rate. A real-time program to quickly identify potential excessive leakage was developed and tested. This program was written in MATLAB© for portability, easy diagnostics and modification. It interfaces directly with EPICS to access the cryogenic system state, and with and NIST REFPROP© for real fluid properties. This program was validated against the actual helium offloaded into the system. The present paper outlines the details of the inventory monitoring program, its validation and a sample of the achieved results.

42 ENGINEERING↗

Addressing Load Imbalance in Bioinformatics and Biomedical Applications: Efficient Scheduling across Multiple GPUs

Computational bioinformatics and biomedical applications frequently contain heterogeneously sized units of work or tasks, for instance due to variability in the sizes of biological sequences and molecules. Variable-sized workloads lead to load imbalances in parallel implementations which detract from efficiency and performance. Many modern computing resources now have multiple graphics processing units(GPUs) per computer for acceleration. These multiple GPU resources need to be used efficiently through balancing of workloads across the GPUs. OpenMP is a portable directive-based parallel programming API used ubiquitously in bioscience applications to program CPUs; recently, the use of OpenMP directives for GPU acceleration has become possible. Here, motivated by experiences with imbalanced loads in GPU-accelerated bioinformatics applications, we address the load balancing problem using OpenMP task-to-GPU scheduling combined with OpenMP GPU offloading for multiply heterogeneous workloads – loads with both variable input sizes, and simultaneously, variable convergence rates for algorithms with a stochastic component – scheduled across multiple GPUs. We aim to develop strategies which are both easy to use and have lower overheads, and may be incorporated incrementally in existing programs which already make use of OpenMP for CPU-based threading in order to make use of multi-GPU computers. We test different combinations of input size variability and convergence rate variability, and characterize the effects of these different scenarios on the performance of scheduling strategies across multiple GPUs with OpenMP. We present several dynamic scheduling solutions for different parallel patterns, explore optimizations, and provide publicly available example computational kernels to make these strategies easy to use in programs. This work will enable application developers to efficiently and easily use multiple GPUs for imbalanced workloads found in bioinformatics and biomedical applications.

Thavappiragasam, Mathialakan↗

RISE: Reducing I/O Contention in Staging-based Extreme-Scale In-situ Workflows

While in-situ workflow formulations have addressed some of the data-related challenges associated with extreme-scale scientific workflows, these workflows involve complex interactions and different modes of data exchange. In the context of increasing system complexity, such workflows present significant resource management challenges, requiring complex cost-performance tradeoffs. This paper presents RISE, an intelligent staging-based data management middleware, which builds on the DataSpaces framework and performs intelligent scheduling of data management operations to reduce I/O contention. In RISE, data are always written immediately to local buffers to reduce the effect of the transfer impact upon application performance. RISE identifies applications’ data access patterns and moves data towards data consumers only when the network is expected to be idle, reducing the impact of asynchronous background data movement upon critical data read/write requests. Here, we experimentally demonstrate that RISE can take advantage of staging nodes to offload data during writes without degrading application data movement performance.

97 MATHEMATICS AND COMPUTING↗

A Framework for Neural Network Inference on FPGA-Centric SmartNICs

FPGA-based SmartNICs offer great potential to significantly improve the performance of high-performance computing and warehouse data processing by tightly coupling support for reconfigurable data-intensive computation with cross-node communication, thereby mitigating the von Neumann bottleneck. Existing work, however, has been generally been limited in that it assumes an accelerator model where kernels are offloaded to SmartNICs, but most control tasks are left to the CPUs. This leads to frequent waiting, inferior performance, and scaling challenges. In this work, we propose a new distributive data-centric computing framework, named FCsN, for reconfigurable SmartNIC-based systems. Through a lightweight task circulation execution model and its implementation architecture, FCsN allows the complete detaching of kernel execution, control logic, system scheduling, and network communication to the SmarNICs. This boosts performance by: (i) avoiding the control dependency with CPUs and (ii) supporting streaming kernel execution and network communication at line rate and in a very fine-grained manner. We demonstrate the efficiency and flexibility of FCsN using various types of neural network applications including graph neural networks; as these last are both irregular and data intensive they offer an especially robust demonstration. Evaluations using commonly-used neural network models and graph datasets show that a system with the support of FCsN can achieve, on average, 144 speedups over the MPI-based standard CPU baselines.

Guo, Anqi↗