Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “computing systems”

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 1,171 records · Page 65

An Application Development Platform for Neuromorphic Computing

Dynamic Adaptive Neural Network Arrays (DANNAs) are neuromorphic computing systems developed as a hardware based approach to the implementation of neural networks. They feature highly adaptive and programmable structural elements, which model arti cial neural networks with spiking behavior. We design them to solve problems using evolutionary optimization. In this paper, we highlight the current hardware and software implementations of DANNA, including their features, functionalities and performance. We then describe the development of an Application Development Platform (ADP) to support efficient application implementation and testing of DANNA based solutions. We conclude with future directions.

Dean, Mark↗

Scalable, Multithreaded, Partially-in-Place Sorting

A recent trend in hardware development is producing computing systems that are stretching the number of cores and size of shared-memory beyond where most fundamental serial algorithms perform well. The expectation is that this trend will continue. So it makes sense to rethink our fundamental algorithms such as sorting. There are many situations where data that needs to be sorted will actually fit into the shared memory so applications could benefit from an efficient parallel sorting algorithm. When sorting large data (at least hundreds of Gigabytes) in a single shared memory, there are two factors that affect the algorithm choice. First, does the algorithm sort in-place? And second, does the algorithm scale well beyond tens of threads? Surprisingly, existing algorithms posses either one of these factors, but not both. We present an approach that gracefully degrades in performance as the amount of available working memory decreases relative to the size of the input.

HPC↗

Co-design of Advanced Architectures for Graph Analytics using Machine Learning

A graph is an excellent way of representing relationships among entities. We can use graph analytics to synthesize and analyze such relational data, and extract relevant features that are useful for various tasks such as machine learning. Considering the crucial role of graph analytics in various domains, it is important and timely to investigate the right hardware configurations that can achieve optimal performance for graph workloads on future high-performance computing systems. Design space exploration studies facilitate the selection of appropriate configurations (e.g. memory) to achieve a desired system performance. Recently, the approach of accelerating graph analytics using persistent non-volatile memory has gained a lot of attention. Traditional system simulators such as Gem5 and NVMain can be used to explore the design space of these advanced memory architectures for graph workloads. However, these simulators are slow in execution thus limiting the efficiency of design space exploration studies. To overcome this challenge, we proposed a machine learning based approach to co-design advanced memory architectures for graph workloads. We tested our approach with DRAM, non-volatile memory, and hybrid memory (DRAM+NVM) using a breadth first search benchmark algorithm. Our results showed the applicability of the proposed machine learning based approach to the co-design of the advanced memory architectures. In this paper, we provide recommendations on selecting advanced memory architectures to achieve desired performance for graph workloads. We also discuss the performances of different machine learning models that were considered in this study.

Kurte, Kuldeep↗

Asynchronous in situ connected-components analysis for complex fluid flows

Fluid flow in porous media is at the heart of problems such as groundwater contamination and carbon sequestration, and presents an important challenge for scientific computing. For this class of problem, large three-dimensional simulations are performed to advance scientific inquiry. On massively parallel computing systems, the volume of data generated by such approaches can become a productivity bottleneck if the raw data generated from the simulation is analyzed in a post-processing step. We present a physics-based framework for in situ data reduction that is theoretically grounded in multiscale averaging theory. We show how task parallelism can be exploited to concurrently perform a variety of analysis tasks with data-dependent costs, including the generation of iso-surfaces, morphological analyses, and connected components analysis. A task management framework is constructed to launch asynchronous analysis threads, manage dependencies between different tasks, promote data locality and hide the impact of data transfers. The framework is applied to analyze GPU-based simulations of two-fluid flow in porous media, generating a set of averaged measures that represents the overall system behavior. We demonstrate how the approach can be applied to perform physically-consistent averaging over fluid subregions using connected components analysis. Simulations performed on Oak Ridge National Lab s Titan supercomputer are profiled to demonstrate the performance of the associated multi-threaded in situ analysis for typical production simulation of two-fluid flow.

Mcclure, James↗

Adaptive Threshold-Based Zonal Isolation of Faults in a Multiterminal DC Using Local Measurements

Fast and accurate methods of fault detection and isolation are a pre-emptive measure in multiterminal dc (MTdc) systems. Zonal isolation of faults is necessary to prevent any misopertation of the dc breakers that can lead to a shutdown of the network. Existing techniques require fast-communication or data synchronization methods have their own disadvantages. This article proposes a method for efficient fault zone isolation without the need of a communication link that prevents any misoperation of the dc breakers in a radial MTdc. This method provides individual local measurement-based control to the hybrid dc circuit breakers (dcCB). Faults created outside the zone of protection for a breaker create a change in the rate of change of current or voltage leading to misoperation. To avoid detection using a fixed threshold, an adaptive threshold based approach is suggested that updates the threshold based on the present operating status. Sensitivity analysis by varying the current limiting inductance and fault location is performed. In conclusion, a three terminal radial model of an MTdc is used for zonal isolation using power system computer aided design/electromagnetic transients including dc.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Putting Regional Climate Prediction in Reach

Developing the capability to project how climate change will affect a particular region--any region--on Earth is a high-performance computing problem with enormous economic and health impacts. Techniques to resolve regional-scale features in global climate models are being developed and validated on leadership computing systems. In time, these regional-scale-resolving global climate models will arm local and state decision makers with the information needed to prepare for both short- and long-term impacts of climate change.

54 ENVIRONMENTAL SCIENCES↗

Evolution at the Edge: Real-Time Evolution for Neuromorphic Engine Control

Neuromorphic computing systems are attractive for real-time control at the edge because of their low power operation, real-time processing capabilities and their potential ability to do online learning. In this work, we describe an approach for performing real-time evolution of spiking neural networks for neuromorphic systems at the edge called Neuromorphic Optimization using Dynamic Evolutionary Systems or NODES. We apply this approach to real-time combustion engine control and develop an engine-specific hardware platform for NODES called FireBox. We demonstrate how the real-time evolution approach works in simulation and the performance of networks trained in simulation on the physical engine.

Maldonado Puente, Bryan [ORNL] (ORCID:000000033880↗

Two-Stage Reinforcement Learning Policy Search for Grid-Interactive Building Control

This paper develops an intelligent grid-interactive building controller, which optimizes building operation during both normal hours and demand response (DR) events. To avoid costly on-demand computation and to adapt to non-linear building models, the controller utilizes reinforcement learning (RL) and makes real-time decisions based on a near-optimal control policy. Learning such a policy typically amounts to solving a hard non-convex optimization problem. We propose to address this problem with a novel global-local policy search method. In the first stage, an RL algorithm based on zero-order gradient estimation is leveraged to search for the optimal policy globally, due to its scalability and the potential to escape some poor performing local optima. The obtained policy is then fine-tuned locally to bring the first-stage solution closer to that of the original unsmoothed problem. Experiments on a simulated five-zone commercial building demonstrate the advantages of the proposed method over existing learning approaches. They also show that the learned control policy outperforms a pragmatic linear model predictive controller (MPC) and approaches the performance of an oracle MPC in testing scenarios. Using a state-of-the-art advanced computing system, we demonstrate that the controller can be learned and deployed within hours of training.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Tutorial: Parallel simulation on supercomputers

This tutorial introduces typical hardware and software characteristics of extant and emerging supercomputing platforms, and presents issues and solutions in executing large-scale parallel discrete-event simulation scenarios on such high performance computing systems. Covered topics include synchronization, model organization, example applications, and observed performance from illustrative large-scale runs.

Perumalla, Kalyan S.↗

Using simulation to quantify the reliability of control software

More than two decades ago, Butler and Finelli examined the problem of experimentally demonstrating the reliability of safety critical software and concluded that it was impractical. We revisit this conclusion in the light of recent advances in computer system virtualization technology and the capability to link virtualization tools to simulation models of physical environments. A specific demonstration of testing for reliability is offered using software that is part of a building control system. Extrapolating the results of this demonstration, we conclude that experimental demonstrations of high reliability may now be feasible for some applications.

Nutaro, James J.↗

In vivo verification of superficial dose for head and neck treatments using intensity-modulated techniques

Skin dose is one of the key issues for clinical dosimetry in radiation therapy. Currently planning computer systems are unable to accurately predict dose in the buildup region, leaving ambiguity as to the dose levels actually received by the patient's skin during radiotherapy. This is one of the prime reasons why in vivo measurements are necessary to estimate the dose in the buildup region. A newly developed metal-oxide-semiconductor-field-effect-transistor (MOSFET) detector designed specifically for dose measurements in rapidly changing dose gradients was introduced for accurate in vivo skin dosimetry. The feasibility of this detector for skin dose measurements was verified in comparison with plane parallel ionization chamber and radiochromic films. The accuracy of a commercial treatment planning system (TPS) in skin dose calculations for intensity-modulated radiation therapy treatment of nasopharyngeal carcinoma was evaluated using MOSFET detectors in an anthropomorphic phantom as well as on the patients. Results show that this newly developed MOSFET detector can provide a minimal but highly reproducible intrinsic buildup of 7 mg cm{sup -2} corresponding to the requirements of personal surface dose equivalent Hp (0.07). The reproducibility of the MOSFET response, in high sensitivity mode, is found to be better than 2% at the phantom surface for the doses normally delivered to the patients. The MOSFET detector agrees well with the Attix chamber and the EBT Gafchromic registered film in terms of surface and buildup region dose measurements, even for oblique incident beams. While the dose difference between MOSFET measurements and TPS calculations is within measurement uncertainty for the depths equal to or greater than 0.5 cm, an overestimation of up to 8.5% was found for the surface dose calculations in the anthropomorphic phantom study. In vivo skin dose measurements reveal that the dose difference between the MOSFET results and the TPS calculations was on average -7.2%, ranging from -4.3% to -9.2%. The newly designed MOSFET detector encapsulated into a thin water protective film has a minimal reproducible intrinsic buildup recommended for skin dosimetry. This feature makes it very suitable for routine IMRT QA and accurate in vivo skin dosimetry.

61 RADIATION PROTECTION AND DOSIMETRY↗

More than 10 years experience of beam monitoring with the Gantry 1 spot scanning proton therapy facility at PSI

Purpose: The beam monitoring equipments developed for the first PSI spot scanning proton therapy facility, Gantry 1, have been successfully used for more than 10 years. The purpose of this article is to summarize the author's experience in the beam monitoring technique for dynamic proton scanning. Methods: The spot dose delivery and verification use two independent beam monitoring and computer systems. In this article, the detector construction, electronic system, dosimetry, and quality assurance results are described in detail. The beam flux monitor is calibrated with a Faraday cup. The beam position monitoring is realized by measuring the magnetic fields of deflection magnets with Hall probes before applying the spot and by checking the beam position and width with an ionization strip chamber after the spot delivery. Results: The results of thimble ionization chamber dosimetry measurements are reproducible (with a mean deviation of less than 1% and a standard deviation of 1%). The resolution in the beam position measurement is of the order of a tenth of a millimeter. The tolerance of the beam position delivery and monitoring during scanning is less than 1.5 mm. Conclusions: The experiences gained with the successful operation of Gantry 1 represent a unique and solid background for the development of a new system, Gantry 2, in order to perform new advanced scanning techniques.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

SU-E-T-794: Validation of the MCNPX Monte Carlo Model of the Ocular Beam Line at the Proton Therapy Center Houston

Purpose: The PTCH is preparing the ocular proton beam nozzle for clinical use. Currently commissioning measurements are being performed using films, diodes and ionization chambers. In parallel, a Monte Carlo model of the beam line was created for integration into the automated Monte Carlo treatment plan computation system, MC{sup 2}. This work aims to compare Monte Carlo predictions to measured proton doses in order to validate the Monte Carlo model. Methods: A complete model of the double scattering ocular beam line has been created and is capable of simulating proton beams with a comprehensive set of beam modifying devices, including eleven different range modulator wheels. Simulations of doses in water were scored and compare to ion chamber measurements of depth doses, lateral dose profiles extracted from half beam block exposures of films, and diode measurements of lateral penumbrae at various depths. Results: All comparison resulted in an average relative entrance dose difference of less than 3% and peak dose difference of less than 2%. All range differences were smaller than 0.2 mm. The differences in the lateral beam profiles were smaller than 0.2 mm, and the differences in the penumbrae were all smaller than 0.4%. Conclusion: All available data shows excellent agreement of simulations and measurements. More measurements will have to be performed in order to completely and systematically validate the model. Besides simulating and measuring PDDs and lateral profiles of all remaining range modulator wheels, the absolute dosimetry factors in terms of number of source protons per monitor unit have to be determined.

60 APPLIED LIFE SCIENCES↗

Trajectory analysis of the rotational dynamics of molecules

A method for analysis of the rotational dynamics of molecular systems has been proposed on the basis of the calculation of the set of exact classical vibrational–rotational trajectories. It has been proposed to compose and to numerically solve the complete system of dynamic equations consisting of Hamilton’s equations and generalized Euler equations for an arbitrary system. The computer algebra system can be applied to automatize the process of derivation and subsequent solution of dynamic equations. The variation of the picture of known bifurcation in the rotational dynamics of symmetric triatomic hydride molecules with an increase in vibrational excitation has been studied within the proposed approach. It has been shown that manifestations of bifurcation completely disappear at a quite high level of vibrational excitations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Enabling event tracing at leadership-class scale through I/O forwarding middleware. In: HPDC '12 Proceedings of the 21st international symposium on High-Performance Parallel and Distributed Computing

Event tracing is an important tool for understanding the performance of parallel applications. As concurrency increases in leadership-class computing systems, the quantity of performance log data can overload the parallel file system, perturbing the application being observed. In this work we present a solution for event tracing at leadership scales. We enhance the I/O forwarding system software to aggregate and reorganize log data prior to writing to the storage system, significantly reducing the burden on the underlying file system for this type of traffic. Furthermore, we augment the I/O forwarding system with a write buffering capability to limit the impact of artificial perturbations from log data accesses on traced applications. To validate the approach, we modify the Vampir tracing toolset to take advantage of this new capability and show that the approach increases the maximum traced application size by a factor of 5x to more than 200,000 processes.

Ilsche, Thomas↗

BCL: A Cross-Platform Distributed Data Structures Library

One-sided communication is a useful paradigm for irregular parallel applications, but most one-sided programming environments, including MPI's one-sided interface and PGAS programming languages, lack application-level libraries to support these applications. We present the Berkeley Container Library, a set of generic, cross-platform, high-performance data structures for irregular applications, including queues, hash tables, Bloom filters and more. BCL is written in C++ using an internal DSL called the BCL Core that provides one-sided communication primitives such as remote get and remote put operations. The BCL Core has backends for MPI, OpenSHMEM, GASNet-EX, and UPC++, allowing BCL data structures to be used natively in programs written using any of these programming environments. Along with our internal DSL, we present the BCL ObjectContainer abstraction, which allows BCL data structures to transparently serialize complex data types while maintaining efficiency for primitive types. We also introduce the set of BCL data structures and evaluate their performance across a number of high-performance computing systems, demonstrating that BCL programs are competitive with hand-optimized code, even while hiding many of the underlying details of message aggregation, serialization, and synchronization.

Brock, Benjamin↗

CommBench: Micro-Benchmarking Hierarchical Networks with Multi-GPU, Multi-NIC Nodes

Modern high-performance computing systems have multiple GPUs and network interface cards (NICs) per node. The resulting network architectures have multilevel hierarchies of subnetworks with different interconnect and software technologies. These systems offer multiple vendor-provided communication capabilities and library implementations (IPC, MPI, NCCL, RCCL, OneCCL) with APIs providing varying levels of performance across the different levels. Understanding this performance is currently difficult because of the wide range of architectures and programming models (CUDA, HIP, OneAPI). We present CommBench, a library with cross-system portability and a high-level API that enables developers to easily build microbenchmarks relevant to their use cases and gain insight into the performance (bandwidth & latency) of multiple implementation libraries on different networks. We demonstrate CommBench with three sets of microbenchmarks that profile the performance of six systems. Our experimental results reveal the effect of multiple NICs on optimizing the bandwidth across nodes and also present the performance characteristics of four available communication libraries within and across nodes of NVIDIA, AMD, and Intel GPU networks.

Hidayetoglu, Mert↗

HP-MDR: High-performance and Portable Data Refactoring and Progressive Retrieval with Advanced GPUs

Scientific applications produce vast amounts of data, posing grand challenges in the underlying data management and analytic tasks. Progressive compression is a promising way to address this problem, as it allows for on-demand data retrieval with significantly reduced data movement cost. However, most existing progressive methods are designed for CPUs, leaving a gap for them to unleash the power of today’s heterogeneous computing systems with GPUs.In this work, we propose HP-MDR, a high-performance and portable data refactoring and progressive retrieval framework for GPUs. Our contributions are four-fold: (1) We carefully optimize the bitplane encoding and lossless encoding, two key stages in progressive methods, to achieve high performance on GPUs; (2) We propose pipeline optimization and incorporate it with data refactoring and progressive retrieval workflows to further enhance the performance for large data process; (3) We leverage our framework to enable high-performance data retrieval with guaranteed error control for common Quantities of Interest; (4) We evaluate HP-MDR and compare it with state of the arts using five real-world datasets. Experimental results demonstrate that HP-MDR delivers an average 13.68 × and 6.31 × throughput in data refactoring and progressive retrieval tasks, respectively. It also leads to 11.22 × throughput for recomposing required data representations under Quantity-of-Interest error control and 6.04 × performance for the corresponding end-to-end data retrieval, when compared with state-of-the-art solutions.

Li, Yanliang [University of Oregon]↗