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 73 records · Page 4

Porting fragmentation methods to GPUs using an OpenMP API: Offloading the resolution-of-the-identity second-order Møller–Plesset perturbation method

Here, using an OpenMP Application Programming Interface, the resolution-of-the-identity second-order Møller–Plesset perturbation (RI-MP2) method has been off-loaded onto graphical processing units (GPUs), both as a standalone method in the GAMESS electronic structure program and as an electron correlation energy component in the effective fragment molecular orbital (EFMO) framework. First, a new scheme has been proposed to maximize data digestion on GPUs that subsequently linearizes data transfer from central processing units (CPUs) to GPUs. Second, the GAMESS Fortran code has been interfaced with GPU numerical libraries (e.g., NVIDIA cuBLAS and cuSOLVER) for efficient matrix operations (e.g., matrix multiplication, matrix decomposition, and matrix inversion). The standalone GPU RI-MP2 code shows an increasing speedup of up to 7.5× using one NVIDIA V100 GPU with one IBM 42-core P9 CPU for calculations on fullerenes of increasing size from 40 to 260 carbon atoms using the 6-31G(d)/cc-pVDZ-RI basis sets. A single Summit node with six V100s can compute the RI-MP2 correlation energy of a cluster of 175 water molecules using the correlation consistent basis sets cc-pVDZ/cc-pVDZ-RI containing 4375 atomic orbitals and 14 700 auxiliary basis functions in ~0.85 h. In the EFMO framework, the GPU RI-MP2 component shows near linear scaling for a large number of V100s when computing the energy of an 1800-atom mesoporous silica nanoparticle in a bath of 4000 water molecules. The parallel efficiencies of the GPU RI-MP2 component with 2304 and 4608 V100s are 98.0% and 96.1%, respectively.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Hybrid programming-model strategies for GPU offloading of electronic structure calculation kernels

To address the challenge of performance portability and facilitate the implementation of electronic structure solvers, we developed the basic matrix library (BML) and Parallel, Rapid O(N), and Graph-based Recursive Electronic Structure Solver (PROGRESS) library. The BML implements linear algebra operations necessary for electronic structure kernels using a unified user interface for various matrix formats (dense and sparse) and architectures (CPUs and GPUs). Focusing on density functional theory and tight-binding models, PROGRESS implements several solvers for computing the single-particle density matrix and relies on BML. In this paper, we describe the general strategies used for these implementations on various computer architectures, using OpenMP target functionalities on GPUs, in conjunction with third-party libraries to handle performance critical numerical kernels. In this study, we demonstrate the portability of this approach and its performance in benchmark problems.

36 MATERIALS SCIENCE↗

OpenACC offloading of the MFC compressible multiphase flow solver on AMD and NVIDIA GPUs

GPUs are the heart of the latest generations of supercomputers. We efficiently accelerate a compressible multiphase flow solver via OpenACC on NVIDIA and AMD Instinct GPUs. Optimization is accomplished by specifying the directive clauses gang vector and collapse. Further speedups of six and ten times are achieved by packing user-defined types into coalesced multidimensional arrays and manual inlining via metaprogramming. Additional optimizations yield seven-times speedup of array packing and thirty-times speedup of select kernels on Frontier. Weak scaling efficiencies of 97% and 95% are observed when scaling to 50% of Summit and 87% of Frontier. Strong scaling efficiencies of 84% and 81% are observed when increasing the device count by a factor of 8 and 16 on V100 and MI250X hardware. The strong scaling efficiency of AMD’s MI250X increases to 92% when increasing the device count by a factor of 16 when GPU-aware MPI is used for communication.

Wilfong, Benjamin↗

Offloading Calculations to Computational Storage Devices: Spark and HDFS [Slides]

The objective is to evaluate the capabilities of multiple CSDs (provided by NDG Systems) using Hadoop Filesystem and Apache Spark. The independent variables are: number of CSDs, 0, 1, 2, 4, or 6; size of dataset, 1 GB, 5 GB, 10 GB; type of dataset, one large file with all of the data, 10 files, 100 files. The dependent variables are: job time; execution time. the constants are operations on the dataset.

97 MATHEMATICS AND COMPUTING↗

Active Response Gravity Offload and Method

A variable gravity field simulator can be utilized to provide three dimensional simulations for simulated gravity fields selectively ranging from Moon, Mars, and micro-gravity environments and/or other selectable gravity fields. The gravity field simulator utilizes a horizontally moveable carriage with a cable extending from a hoist. The cable can be attached to a load which experiences the effects of the simulated gravity environment. The load can be a human being or robot that makes movements that induce swinging of the cable whereby a horizontal control system reduces swinging energy. A vertical control system uses a non-linear feedback filter to remove noise from a load sensor that is in the same frequency range as signals from the load sensor.

Dungan, Larry K.↗

Route Optimization for Offloading Congested Meter Fixes

The Optimized Route Capability (ORC) concept proposed by the FAA facilitates traffic managers to identify and resolve arrival flight delays caused by bottlenecks formed at arrival meter fixes when there exists imbalance between arrival fixes and runways. ORC makes use of the prediction capability of existing automation tools, monitors the traffic delays based on these predictions, and searches the best reroutes upstream of the meter fixes based on the predictions and estimated arrival schedules when delays are over a predefined threshold. Initial implementation and evaluation of the ORC concept considered only reroutes available at the time arrival congestion was first predicted. This work extends previous work by introducing an additional dimension in reroute options such that ORC can find the best time to reroute and overcome the 'firstcome- first-reroute' phenomenon. To deal with the enlarged reroute solution space, a genetic algorithm was developed to solve this problem. Experiments were conducted using the same traffic scenario used in previous work, when an arrival rush was created for one of the four arrival meter fixes at George Bush Intercontinental Houston Airport. Results showed the new approach further improved delay savings. The suggested route changes from the new approach were on average 30 minutes later than those using other approaches, and fewer numbers of reroutes were required. Fewer numbers of reroutes reduce operational complexity and later reroutes help decision makers deal with uncertain situations.

Optimized Route Capability↗

BioBot: Innovative Offloading of Astronauts for More Effective Exploration

The BioBot concept consists of a robotic rover which is capable of traversing the same terrain as a spacesuited human. It carries the primary life support system for the astronaut, including consumables, atmosphere revitalization systems (e.g., CO2 scrubbing, humidity and temperature management, ventilation fan), power system (e.g., battery, power management and distribution),and thermal control system (e.g., water sublimator, cooling water pump), along with umbilical lines to connect to the supported astronaut. Although not technically part of life support, it would be logical for the BioBot to also provide long-range communications, video monitoring, tool and sample transport, and other functions to enable and enhance EVA productivity in planetary surface exploration.The design reference scenario for this concept is that astronauts involved in future lunar or Mars exploration will be on the surface for weeks or months rather than days, and will be involved in regular EVA operations. It is not unreasonable to think of geologists spending several days inEVA exploration each week over a prolonged mission duration, with far more ambitious operational objectives than were typical of Apollo. In this scenario, each astronaut will be accompanied by a "BioBot", which will transport their life support system and consumables, an extended umbilical and umbilical reel, and robotic systems capable of controlling the position and motion of the umbilical. The astronaut will be connected to the robot via the umbilical, carrying only a small emergency open-loop life support system similar to those contained in every PLSS. The robotic mobility base will be designed to be capable of traveling anywhere the astronaut can walk, and will also be useful as a transport for the EVA tools, science instrumentation, and collected samples. In addition, the BioBot can potentially carry the astronaut on traverses as well. Such a system will also be a significant enhancement to public engagement in these future exploration missions, as the robotic vehicles can also support high-resolution cameras and high bandwidth communications gear to providehigh-definition video coverage of each crew throughout each EVA sortie.

Akin, David L.↗