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 37 records · Page 2

Development of ARGOS Offloading Assessments and Methodology for Lunar EVA Simulations

The Active Response Gravity Offload System (ARGOS) at NASA Johnson Space Center (JSC) is an analog environment that can offload pressurized suited subjects at various gravity levels. The suit is suspended from a robotic overhead crane by a cable connected to the suit via a gimbal with an adjustable pivot point. There has been increased interest in providing lunar pressurized suited training at ARGOS in preparation for lunar missions. Determination of the appropriate gimbal pivot point location for a given subject is vital for a high-fidelity and functional lunar simulation. Interactions between the pivot point location and center of gravity (CG) can result in righting moments that may lead to artificially stable or unrealistically challenging configurations. Changing the pivot point location is time consuming and repeated adjustment can result in significant loss of valuable pressurized suited time. This paper aims to share the knowledge obtained from the offloading characterization efforts during pressurized suited testing at ARGOS and document the ongoing process to define an appropriate pivot point location through iterative quantitative and qualitative assessments. The human-spacesuit CG locations for the ARGOS lunar simulation were estimated using a 3D body scan and density model combined with spacesuit hardware CAD and specifications. Early pilot testing of the gimbal revealed that setting the pivot point coincident with the modeled CG location was not always possible due to the current gimbal design, and small shifts forward and aft had noticeable effects on subject stability. Fifteen subjects performed a series of CG-related tasks in the xEMU spacesuit to assess simulation acceptability. Through iterative testing, this task list evolved to streamline the process needed to efficiently identify a suitable pivot point for a given subject. The developed methodology will be critical to determine pivot point selection for astronaut training in the xEMU ARGOS environment.

Sarah L. Jarvis↗

Runtime performance of a GAMESS quantum chemistry application offloaded to GPUs

Summary Computational chemistry is at the forefront of solving urgent societal problems, such as polymer upcycling and carbon capture. The complexity of modeling these processes at appropriate length and time scales is mainly manifested in the number and types of chemical species involved in the reactions and may require models of several thousand atoms and large basis sets to accurately capture the chemical complexity and heterogeneity in the physical and chemical processes. The quantum chemistry package General Atomic and Molecular Electronic Structure System (GAMESS) has a wide array of methods that can efficiently and accurately treat complex chemical systems. In this work, we have used the GAMESS Effective Fragment Molecule Orbital (EFMO) method for electronic structure calculation of a challenging mesoporous silica nanoparticle (MSN) model surrounded by about 4700 water molecules to investigate the strong scaling and GPU offloading on hybrid CPU‐GPU nodes. Experiments were performed on the Perlmutter platform at the National Energy Research Scientific Computing Center. Good strong scaling and load balancing have been observed on up to 88 hybrid nodes for different settings of the execution parameters for the calculation considered here. When GPUs are oversubscribed by offloading work from multiple CPU processes, using the NVIDIA multi‐process service (MPS) has consistently reduced time to solution and energy consumed. Additionally, for some configuration parameter settings, oversubscription with MPS improved performance by up to 5.8% over the case without oversubscription.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Benchmarking and Evaluating Unified Memory for OpenMP GPU Offloading

Here, the latest OpenMP standard offers automatic device offloading capabilities which facilitate GPU programming. Despite this, there remain many challenges. One of these is the unified memory feature introduced in recent GPUs. GPUs in current and future HPC systems have enhanced support for unified memory space. In such systems, CPU and GPU can access each other's memory transparently, that is, the data movement is managed automatically by the underlying system software and hardware. Memory over subscription is also possible in these systems. However, there is a significant lack of knowledge about how this mechanism will perform, and how programmers should use it. We have modified several benchmarks codes, in the Rodinia benchmark suite, to study the behavior of OpenMP accelerator extensions and have used them to explore the impact of unified memory in an OpenMP context. We moreover modified the open source LLVM compiler to allow OpenMP programs to exploit unified memory. The results of our evaluation reveal that, while the performance of unified memory is comparable with that of normal GPU offloading for benchmarks with little data reuse, it suffers from significant overhead when GPU memory is over subcribed for benchmarks with large amount of data reuse. Based on these results, we provide several guidelines for programmers to achieve better performance with unified memory.

97 MATHEMATICS AND COMPUTING↗

Lunar Surface Cargo Offloading Concepts

A sustainable presence on the lunar surface will serve as a vital training ground and technology demonstration test site in preparation for future human missions to Mars. Robotic lunar surface campaigns will focus on the exploration of resources providing information on the availability and extraction of usable resources, such as oxygen and water, and prepare the surface for a sustained human presence. Landers, outfitted with sensor packages, will be used to conduct risk-reduction activities and aid in the development of technologies prior to the crewed lunar missions that drive the need to for a logistics supply chain that requires offloading. A series of landers will be required on other planetary surfaces to build up the capabilities, capitalizing on those resources, required for sustained human presence. In each of those landers will be cargo including ascent vehicles, habitats, supplies, science packages, spare parts, fluids commodities for fuel and life support, and others varying in volume and ranging from mass in hundreds of kilograms to an estimated 6-14 metric tons to support Human Landing Systems and surface logistics requirements. This paper will examine the challenge of offloading examples of these cargo elements from different categories of landers on the lunar surface using a variety of methodologies. Challenges on the lunar surface arise with the conditions present (thermal, lighting, communications, regolith consistency), the desire to minimize mass of all landed systems, the desire to perform much of the activities with limited to minimal human interaction, and the overall configuration of the landers that are responsible for landing the cargo.

Lunar↗

Lunar Surface Cargo Offloading Concepts

A sustainable presence on the lunar surface will serve as a vital training ground and technology demonstration test site in preparation for future human missions to Mars. Robotic lunar surface campaigns will focus on the exploration of resources providing information on the availability and extraction of usable resources, such as oxygen and water, and prepare the surface for a sustained human presence. Landers, outfitted with sensor packages, will be used to conduct risk-reduction activities and aid in the development of technologies prior to the crewed lunar missions that drive the need to for a logistics supply chain that requires offloading. A series of landers will be required on other planetary surfaces to build up the capabilities, capitalizing on those resources, required for sustained human presence. In each of those landers will be cargo including ascent vehicles, habitats, supplies, science packages, spare parts, fluids commodities for fuel and life support, and others varying in volume and ranging from mass in hundreds of kilograms to an estimated 6-14 metric tons to support Human Landing Systems and surface logistics requirements. This paper will examine the challenge of offloading examples of these cargo elements from different categories of landers on the lunar surface using a variety of methodologies. Challenges on the lunar surface arise with the conditions present (thermal, lighting, communications, regolith consistency), the desire to minimize mass of all landed systems, the desire to perform much of the activities with limited to minimal human interaction, and the overall configuration of the landers that are responsible for landing the cargo.

Lunar↗

Just-in-Time Compilation and Link-Time Optimization for OpenMP Target Offloading

Following the mass adoption of external accelerators for high performance computing, the overall performance of many applications has become increasingly dependent on relatively small accelerated kernels. As static analysis is fundamentally limited by dynamic values and external definitions, standard ahead-of-time compilation is not always sufficient to achieve the best performance. Furthermore, many users looking to port an existing application to run on an external accelerator will not want to fundamentally restructure their programs. These and other problems can be addressed through both link-time optimization (LTO) and just-in-time (JIT) compilation, but until now had sparse and inconsistent support from the compiler. In this work, we present a new compilation method that enables device-side LTO as well as a transparent JIT compilation tool-chain for OpenMP target offloading. Our contributions include an entirely new device linking and embedding scheme to enable LTO as well as a novel JIT engine to efficiently optimize OpenMP offloading regions at run-time. We also introduce a persistent caching system to improve end-to-end runtime using the JIT engine and minimize kernel launching overheads. We measure the performance of our LTO and JIT implementation via several real-world scientific applications. With our optimizations we observe significant improvements through LTO on large applications as well as significant end-to-end execution time improvement using JIT.

Tian, Shilei↗

Experimental Characterization of OpenMP Offloading Memory Operations and Unified Shared Memory Support

The OpenMP specification recently introduced support for unified shared memory, allowing implementation to leverage underlying system software to provide a simpler GPU offloading model where explicit mapping of variables is optional. Support for this feature is becoming more available in different OpenMP implementations on several hardware platforms. A deeper understanding of the different implementation’s execution profile and performance is crucial for applications as they consider the performance portability implications of adopting a unified memory offloading programming style. This work introduces a benchmark tool to characterize unified memory support in several OepnMP compilers and runtimes, with emphasis on identifying discrepancies between different OpenMP implementations as to how they various memory allocation strategies interact with unified shared memory. The benchmark tool is used to characterize OpenMP compilers on three leading High Performance Computing platforms supporting different CPU and device architectures. The benchmark tool is used to assess the impact of enabling unified shared memory on the performance of memory-bound code, highlighting implementation differences that should be accounted for when applications consider performance portability across platforms and compilers.

Elwasif, Wael↗

DoCeph: DPU-Offloaded Messaging in Ceph for Reduced Host CPU Utilization

Ceph is a widely used distributed object store, but its messenger layer imposes substantial CPU overhead on the host. To address this limitation, we propose DoCeph, a DPU-offloaded storage architecture for Ceph that disaggregates the system by offloading the communication-intensive messaging component to the DPU while retaining the storage backend on the host. The DPU efficiently manages communication, using lightweight RPC for metadata operations and DMA for data transfer. Moreover, DoCeph introduces a pipelining technique that overlaps data transmission with buffer preparation, mitigating hardware-imposed transfer size limitations. We implemented DoCeph on a Ceph cluster with NVIDIA BlueField-3 DPUs. Evaluation results indicate that DoCeph cuts host CPU usage by up to 92% while sustaining stable throughput and providing larger performance benefits for object writes over 1 MB.

Park, Kuri [Sogang University]↗

Numerical eigen-spectrum slicing, accurate orthogonal eigen-basis, and mixed-precision eigenvalue refinement using OpenMP data-dependent tasks and accelerator offload

Performing a variety of numerical computations efficiently and, at the same time, in a portable fashion requires both an overarching design followed by a number of implementation strategies. All of these are exemplified below as we present transitioning the PLASMA numerical library from relying on dependence-driven large tasks to achieving utilization of fine grain tasking and offload to hardware accelerators while keeping its core dependence sets: OpenMP source code pragmas and runtime for most system-level functionality and basic low-level numerical kernels provided directly by hardware vendors or open source projects with vendor contributions. We also present new algorithmic methods and their efficient parallel implementations including fine grained tasking for eigen-spectrum slicing and offload for mixed-precision eigenvalue refinement. We provide performance, scaling, and numerical results showing sizable gains over the available solutions from either the open source and vendor-provided packages.

Luszczek, Piotr↗

Offloading techniques for large deployable space structures

The validation and verification of large deployable space structures are continual challenges which face the integration and test engineer today. Spar Aerospace Limited has worked on various programs in which such structure validation was required and faces similar tasks in the future. This testing is reported and the different offloading and deployment methods which were used, as well as the proposed methods which will be used on future programs, are described. Past programs discussed include the Olympus solar array ambient and thermal vacuum deployments, and the Anik-E array and reflector deployments. The proposed MSAT reflector and boom ambient deployment tests, as well as the proposed RADARSAT Synthetic Aperture Radar (SAR) ambient and thermal vacuum deployment tests will also be presented. A series of tests relating to various component parts of the offloading equipment systems was required. These tests included the characterization and understanding of linear bearings and large (180 in-lbf) constant force spring motors in a thermal vacuum environment, and the results from these tests are presented.

Caravaggio, Levino↗

Test Frame for Gravity Offload Systems

Advances in space telescope and aperture technology have created a need to launch larger structures into space. Traditional truss structures will be too heavy and bulky to be effectively used in the next generation of space-based structures. Large deployable structures are a possible solution. By packaging deployable trusses, the cargo volume of these large structures greatly decreases. The ultimate goal is to three dimensionally measure a boom's deployment in simulated microgravity. This project outlines the construction of the test frame that supports a gravity offload system. The test frame is stable enough to hold the gravity offload system and does not interfere with deployment of, or vibrations in, the deployable test boom. The natural frequencies and stability of the frame were engineered in FEMAP. The test frame was developed to have natural frequencies that would not match the first two modes of the deployable beam. The frame was then modeled in Solidworks and constructed. The test frame constructed is a stable base to perform studies on deployable structures.

Murray, Alexander R.↗

Active Response Gravity Offload System

The Active Response Gravity Offload System (ARGOS) provides the ability to simulate with one system the gravity effect of planets, moons, comets, asteroids, and microgravity, where the gravity is less than Earth fs gravity. The system works by providing a constant force offload through an overhead hoist system and horizontal motion through a rail and trolley system. The facility covers a 20 by 40-ft (approximately equals 6.1 by 12.2m) horizontal area with 15 ft (approximately equals4.6 m) of lifting vertical range.

Valle, Paul↗

Development and Evaluation of the Active Response Gravity Offload System as a Lunar and Martian EVA Simulation Environment

In preparation for future exploration missions, NASA seeks the ability to simulate partial-gravity operations for use in ground-based research, crew training, and engineering design evaluations. The Active Response Gravity Offload System (ARGOS) at the Johnson Space Center (JSC) is designed to simulate reduced gravity environments, such as lunar, Martian, or microgravity, using a robotic system similar to an overhead bridge crane. ARGOS continuously offloads a portion of a suited human’s weight during all dynamic motions within the test facility, which can include basic functional movements such as walking, running, and jumping, as well as a wide range of planetary surface activities. This system will be used as part of a metabolic-rate task characterization study to determine the workload associated with partial-gravity extravehicular activity (EVA). Pilot testing was conducted using the MKIII prototype planetary space suit and two gimbal designs to determine the ability of the ARGOS test environment to simulate planetary EVA operations. This paper will describe the lessons learned from the feasibility testing, simulation-environment mockup design, and the results from the pilot tests and their influence into the final study design. Being able to effectively simulate partial-gravity environments and characterize the performance of crewmembers will have an impact on multiple domains including suit design, task design, thermal models, and life-support-system capacity verification plans, among others.

Omar S Bekdash↗

LSMS–L35, Miniature Crane for Payload Offloading and Manipulation: Development, and Application

The Lightweight Surface Manipulation System (LSMS) is a robotic agent for autonomous surface construction activities on planetary surfaces, that was designed at NASA Langley Research Center and has over a decade of research and development. The LSMS is a key component to achieving many goals of the NASA Artemis program. The LSMS is lightweight, structurally efficient system that can be easily packaged for launch and deployment on-surface, capable of a suite of surface activities enabled by modular end-effectors at the wrist. The focus of recent development work has been on using the LSMS for payload offloading and handling from lunar landers. Discussed in the paper is the development of the LSMS-L35 hardware (35 kg wrist lifting capacity on the lunar surface), designed to integrate with a Commercial Lunar Payload Services (CLPS) lander to offload payloads to the surface. The LSMS-L35 hardware development is part of a larger effort to enable autonomous payload handling and manipulation.

Iok M. Wong↗

LANDO: Developing Autonomous Payload Offloading Capabilities for Lunar Surface Operations

Introduction: The Lightweight Surface Manipulation System (LSMS) AutoNomy capabilities Development for surface Operations and construction (LANDO) project is an Early Career Initiative selected for funding by NASA Space Technology Mission Directorate. LANDO is developing a general-purpose autonomy framework applicable to serial and tension-actuated manipulation agents, that will be validated using an existing prototype of the LSMS-L35 (35-kg wrist lift capacity on the lunar surface [Fig. 1], sized for a Commercial Lunar Pay-load Services (CLPS) mission). The autonomous LSMS-L35 will be used to demonstrate autonomous payload handling capabilities for Lunar and other planetary surfaces, directly addressing STMD capability gaps in autonomous excavation and construction operations, advanced robotics and spacecraft autonomy technologies, and technologies supporting emerging space industries including the In-Space Servicing, Assembly and Manufacturing national strategy. LSMS: The LSMS is a tension actuated robotic agent that is scalable (reach and lifting capacity in different gravity environments), versatile (types of surface operations), and reusable. Compared to serial arms, the LSMS provides significantly higher structural efficiency and mechanical advantage, enabling a greater payload lift capacity at a lower system mass. The LSMS is envisioned to be a crucial part of the excavation and construction portfolio, capable of supporting a variety of activities on the lunar surface. Autonomous payload handling is one of the first activities the LSMS can support that develops capabilities that are extensible to other surface operations. Payload handling is required to: remove payloads from a lander; place payloads on mobile agents for transport from a lander to construction site/assembly point; emplace payloads in their operational configuration, and aggregate components to create an asset. As an example of this critical gap, manifested CLPS missions do not currently have a ubiquitous payload offloading capability; payloads (excluding rovers) are designed to remain on the lander. Why Autonomy? Autonomous robotic systems capable of carrying out excavation and construction operations are a fundamental and critical capability required for realizing the NASA Artemis program vision to “emplace and build the infrastructure, systems, and robotic missions that can enable a sustained lunar surface presence.” While teleoperation is still feasible for lunar surface operations, increased latency at Mars will require validated supervised autonomous technologies capable of operating with minimal human involvement (human-on-the-loop) unless an unexpected event occurs requiring human intervention. Autonomy reduces operator burden, allows operations to continue during uncrewed periods, in-creases the safety of operations by automatically detecting and handling faults, and allows operating in high latency environments. Development Activities: LANDO is extending critical autonomous operations to the manipulation domain and creating an integrated system, based on reusable software modules, that is capable of planning and executing payload handling and autonomous surface operations without requiring hu-man intervention beyond a supervisory role. The priority features under development are 1) autonomously offload payloads from a tilted lander deck without buckling the LSMS; 2) sensing whether a payload is safe to lift and handle; and 3) integrate with Astrobotic’s CLPS lander. The poster presentation will highlight current development activities over the past year on LSMS-L35 prototype hard-ware design, and autonomy software.

in-space assembly↗

ATHLETE Offloader Limb as a High-capacity Crane

A new concept for the NASA Jet Propulsion Laboratory (JPL) All-Terrain Hex-Limbed Extra-Terrestrial Explorer (ATHLETE) robotic constructor / mobility system employs tendon-driven actuation of individual limbs, similar to high-capacity cranes in terrestrial work environments, that can offload cargo from tall landers. While maintaining mechanical joints that allow each limb to function as a highly dexterous multi-Degree-Of-Freedom (DOF) robotic arm, tendon-driven truss sections increase the capacity of moment loads for extended configurations of the limb. The tendon-driven crane-like limb will be capable of offloading cargo from tall landers such as the SpaceX Starship. This paper describes the structural analysis, operations, calibration, and actuation of limbs designed to address specific target load cases that might be required for human exploration missions on planetary surfaces.

Wilcox, Brian↗

ConnectX-2 InfiniBand Management Queues: First Investigation of the New Support for Network Offloaded Collective Operations

This paper introduces the newly developed Infini-Band (IB) Management Queue capability, used by the Host Channel Adapter (HCA) to manage network task data flow dependancies, and progress the communications associated with such flows. These tasks include sends, receives, and the newly supported wait task, and are scheduled by the HCA based on a data dependency description provided by the user. This functionality is supported by the ConnectX-2 HCA, and provides the means for delegating collective communication management and progress to the HCA, also known as collective communication offload. This provides a means for overlapping collective communications managed by the HCA and computation on the Central Processing Unit (CPU), thus making it possible to reduce the impact of system noise on parallel applications using collective operations. This paper further describes how this new capability can be used to implement scalable Message Passing Interface (MPI) collective operations, describing the high level details of how this new capability is used to implement the MPI Barrier collective operation, focusing on the latency sensitive performance aspects of this new capability. This paper concludes with small scale benchmark experiments comparing implementations of the barrier collective operation, using the new network offload capabilities, with established point-to-point based implementations of these same algorithms, which manage the data flow using the central processing unit. These early results demonstrate the promise this new capability provides to improve the scalability of high performance applications using collective communications. The latency of the HCA based implementation of the barrier is similar to that of the best performing point-to-point based implementation managed by the central processing unit, starting to outperform these as the number of processes involved in the collective operation increases.

Graham, Richard L↗

GPU-enabled extreme-scale turbulence simulations: Fourier pseudo-spectral algorithms at the exascale using OpenMP offloading

Fourier pseudo-spectral methods for nonlinear partial differential equations are of wide interest in many areas of advanced computational science, including direct numerical simulation of three-dimensional (3-D) turbulence governed by the Navier-Stokes equations in fluid dynamics. This paper presents a new capability for simulating turbulence at a new record resolution up to 35 trillion grid points, on the world's first exascale computer, Frontier, comprising AMD MI250x GPUs with HPE's Slingshot interconnect and operated by the US Department of Energy's Oak Ridge Leadership Computing Facility (OLCF). Key programming strategies designed to take maximum advantage of the machine architecture involve performing almost all computations on the GPU which has the same memory capacity as the CPU, performing all-to-all communication among sets of parallel processes directly on the GPU, and targeting GPUs efficiently using OpenMP offloading for intensive number-crunching including 1-D Fast Fourier Transforms (FFT) performed using AMD ROCm library calls. With 99% of computing power on Frontier being on the GPU, leaving the CPU idle leads to a net performance gain via avoiding the overhead of data movement between host and device except when needed for some I/O purposes. Memory footprint including the size of communication buffers for MPI_ALLTOALL is managed carefully to maximize the largest problem size possible for a given node count. Detailed performance data including separate contributions from different categories of operations to the elapsed wall time per step are reported for five grid resolutions, from 2048 3 on a single node to 32768 3 on 4096 or 8192 nodes out of 9408 on the system. Both 1D and 2D domain decompositions which divide a 3D periodic domain into slabs and pencils respectively are implemented. The present code suite (labeled by the acronym GESTS, GPUs for Extreme Scale Turbulence Simulations) achieves a figure of merit (in grid points per second) exceeding goals set in the Center for Accelerated Application Readiness (CAAR) program for Frontier. The performance attained is highly favorable in both weak scaling and strong scaling, with notable departures only for 2048 3 where communication is entirely intra-node, and for 32768 3 , where a challenge due to small message sizes does arise. Communication performance is addressed further using a lightweight test code that performs all-to-all communication in a manner matching the full turbulence simulation code. Performance at large problem sizes is affected by both small message size due to high node counts as well as dragonfly network topology features on the machine, but is consistent with official expectations of sustained performance on Frontier. Overall, although not perfect, the scalability achieved at the extreme problem size of 32768 3 (and up to 8192 nodes — which corresponds to hardware rated at just under 1 exaflop/sec of theoretical peak computational performance) is arguably better than the scalability observed using prior state-of-the-art algorithms on Frontier's predecessor machine (Summit) at OLCF. New science results for the study of intermittency in turbulence enabled by this code and its extensions are to be reported separately in the near future.

3D fast Fourier transform↗