Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “performance tuning”

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 469 records · Page 26

Online transfer learning strategy for enhancing the scalability and deployment of deep reinforcement learning control in smart buildings

In recent years, advanced control strategies based on Deep Reinforcement Learning (DRL) proved to be effective in optimizing the management of integrated energy systems in buildings, reducing energy costs and improving indoor comfort conditions when compared to traditional reactive controllers. However, the scalability and implementation of DRL controllers are still limited since they require a considerable amount of time before converging to a near-optimal solution. This issue is currently addressed in literature through the offline pre-training of the DRL agent. However this solution results in two main critical issues: (1) the need to develop a building surrogate model to perform the training task, and (2) the need to perform a fine-tuning process over several training episodes to obtain a near-optimal control policy. In this context, this paper introduces an Online Transfer Learning (OTL) strategy that exploits two knowledge-sharing techniques, weight-initialization and imitation learning, to transfer a DRL control policy from a source office building to various target buildings in a simulation environment coupling EnergyPlus and Python. A DRL controller based on discrete Soft Actor–Critic (SAC) is trained on the source building to manage the operation of a cooling system consisting of a chiller and a thermal storage. Several target buildings are defined to benchmark the performance of the OTL strategy with that of a Rule-Based Controller (RBC) and two DRL-based control strategies, deployed in offline and online fashion. The strategy adopted for OTL emulates the real world implementation with a simulation process by implementing the transferred DRL agent for a single episode in the target buildings. Target buildings have the same geometrical features and are served by the same energy system as the source building, but differ in terms of weather conditions, electricity price schedules, occupancy patterns, and building envelope efficiency levels. The results show that the OTL strategy can reduce the cumulated sum of temperature violations on average by 50% and 80% respectively when compared to RBC and online DRL while enhancing the energy system operation with electricity cost savings ranging between 20% and 40%. Furthermore, the OTL agent performs slightly worse than the offline DRL controller but it does not require any modeling effort and can be implemented directly on target buildings emulating a real-world implementation.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Matrix Factorizations at Scale: a Comparison of Scientific Data Analytics in Spark and C+MPI Using Three Case Studies

We explore the trade-offs of performing linear algebra using Apache Spark, compared to traditional C and MPI implementations on HPC platforms. Spark is designed for data analytics on cluster computing platforms with access to local disks and is optimized for data-parallel tasks. We examine three widely-used and important matrix factorizations: NMF (for physical plausibility), PCA (for its ubiquity) and CX (for data interpretability). We apply these methods to 1.6TB particle physics, 2.2TB and 16TB climate modeling and 1.1TB bioimaging data. The data matrices are tall-and-skinny which enable the algorithms to map conveniently into Spark’s data parallel model. We perform scaling experiments on up to 1600 Cray XC40 nodes, describe the sources of slowdowns, and provide tuning guidance to obtain high performance.

Gittens, Alex↗

Optimization and understanding of ZnO nanoarray supported Cu-ZnO-Al 2 O 3 catalyst for enhanced CO 2 -methanol conversion at low temperature and pressure

Cu-ZnO-Al 2 O 3 is the most widely applied catalyst for CO 2 hydrogenation to methanol. However, it is still a challenge to produce methanol using this catalyst under low-temperature (<250 °C) and low-pressure (<10 bar) conditions with desirable yield and selectivity. In this work, by tuning the experimental processing parameters such as solvent, loading amount, and annealing temperature, highly improved ZnO nanoarray supported Cu-ZnO-Al 2 O 3 catalysts have been successfully demonstrated. Here, by using organic solvent (N,N-dimethylformamide (DMF), acetone, or isopropanol) for dip-coating loading process instead of deionized (DI) water, Cu-ZnO-Al 2 O 3 nanocatalysts was comparatively better dispersed on the nanorod array support with populated and abundant active sites, thus enhancing the methanol yield. With the control of the loading amount and annealing temperature, finely distributed Cu nanoparticles were obtained on the ZnO nanorod surfaces to enhance the interactions between Cu and ZnO nanorod surfaces. Further improvement of the catalyst performance is demonstrated by tuning the reaction space velocity. At 200 °C and 10 bar conditions, the optimized catalyst achieved a methanol yield of 6.46 mol h –1 kg –1 with 100 % selectivity. The good stability after prolonged testing of the catalysts demonstrates the potential practical implementation. The in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) measurements under the 1 bar reveal that the CO 2 hydrogenation to methanol on the ZnO nanoarray supported Cu-ZnO-Al 2 O 3 catalyst follows the CO reaction pathway, due to the surface oxygen vacancies on ZnO nanorods which facilitate CO 2 dissociation.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Autotuning in High-Performance Computing Applications

Autotuning refers to the automatic generation of a search space of possible implementations of a computation that are evaluated through models and/or empirical measurement to identify the most desirable implementation. Autotuning has the potential to dramatically improve the performance portability of petascale and exascale applications. To date, autotuning has been used primarily in high-performance applications through tunable libraries or previously tuned application code that is integrated directly into the application. This paper draws on the authors' extensive experience applying autotuning to high-performance applications, describing both successes and future challenges. If autotuning is to be widely used in the HPC community, researchers must address the software engineering challenges, manage configuration overheads, and continue to demonstrate significant performance gains and portability across architectures. In particular, tools that configure the application must be integrated into the application build process so that tuning can be reapplied as the application and target architectures evolve.

97 MATHEMATICS AND COMPUTING↗

A multi-backend autotuning study of feature selection on GPUs

Abstract Feature selection is an important step in machine learning that can benefit from GPU acceleration. As the number of GPU vendors increases, it is imperative to adapt algorithms such as the minimum Redundancy Maximum Relevance (mRMR) feature selection method to different backends that support several GPU architectures. This work presents a multi-backend implementation of mRMR across CUDA, HIP, and SYCL, and studies its performance when combined with Bayesian optimization and transfer learning to automatically tune execution parameters for different platforms and datasets. Our experimental results show that when tuned, CUDA and HIP achieve comparable performance on NVIDIA architectures, while SYCL exhibits a moderate performance gap. Overall, this work highlights the impact of backend choice and autotuning on GPU-accelerated feature selection and provides insights into deploying mRMR across heterogeneous environments.

Beceiro, Bieito (ORCID:0000000333014890)↗

Optimization and Portability of a Fusion OpenACC-based FORTRAN HPC Code from NVIDIA to AMD GPUs

NVIDIA has been the main provider of GPU hardware in HPC systems for over a decade. Most applications that benefit from GPUs have thus been developed and optimized for the NVIDIA software stack. Recent exascale HPC systems are, however, introducing GPUs from other vendors, e.g. with the AMD GPU-based OLCF Frontier system just becoming available. AMD GPUs cannot be directly accessed using the NVIDIA software stack, and require a porting effort by the application developers. This paper provides an overview of our experience porting and optimizing the CGYRO code, a widely-used fusion simulation tool based on FORTRAN with OpenACC-based GPU acceleration. While the porting from the NVIDIA compilers was relatively straightforward using the CRAY compilers on the AMD systems, the performance optimization required more fine-tuning. In the optimization effort, we uncovered code sections that had performed well on NVIDIA GPUs, but were unexpectedly slow on AMD GPUs. After AMD-targeted code optimizations, performance on AMD GPUs has increased to meet our expectations. Modest speed improvements were also seen on NVIDIA GPUs, which was an unexpected benefit of this exercise.

Sfiligoi, Igor↗

Optimizing Input/Output Using Adaptive File System Policies

Parallel input/output characterization studies and experiments with flexible resource management algorithms indicate that adaptivity is crucial to file system performance. In this paper we propose an automatic technique for selecting and refining file system policies based on application access patterns and execution environment. An automatic classification framework allows the file system to select appropriate caching and pre-fetching policies, while performance sensors provide feedback used to tune policy parameters for specific system environments. To illustrate the potential performance improvements possible using adaptive file system policies, we present results from experiments involving classification-based and performance-based steering.

Madhyastha, Tara M.↗

Cross-Layer Self-Adaptive/Self-Aware System Software for Exascale Systems

The extreme level of parallelism coupled with the limited available power budget expected in the exascale era brings unprecedented challenges that demand optimization of performance, power and resiliency in unison. Scalability on such systems is of paramount importance, while power and reliability issues may change the execution environment in which a parallel application runs. To solve these challenges exascale systems will require an introspective system software that combines system and application observations across all system stack layers with online feedback and adaptation mechanisms. In this paper we propose the design of a novel self-aware, self-adaptive system software in which a kernel-level Monitor, which continuously inspects the evolution of the target system through observation of Sensors, is combined with a user-level Controller, which reacts to changes in the execution environment, explores opportunities to increase performance, save power and adapts applications to new execution scenarios. We show that the monitoring system accurately monitors the evolution of parallel applications with a runtime overhead below 1-2%. As a test case, we design and implement a user-runtime system that aims at optimizing application’s performance and system power consumption on complex hierarchical architectures. Our results show that our adaptive system reaches 98% of performance efficiency of manually-tuned applications.

High Performance computing, Self-aware/self-adapti↗

An analytical method of estimating turbine performance

A method is developed by which the performance of a turbine over a range of operating conditions can be analytically estimated from the blade angles and flow areas. In order to use the method, certain coefficients that determine the weight flow and the friction losses must be approximated. The method is used to calculate the performance of the single-stage turbine of a commercial aircraft gas-turbine engine and the calculated performance is compared with the performance indicated by experimental data. For the turbine of the typical example, the assumed pressure losses and the tuning angles give a calculated performance that represents the trends of the experimental performance with reasonable accuracy. The exact agreement between analytical performance and experimental performance is contingent upon the proper selection of a blading-loss parameter.

Kochendorfer, Fred D↗

Domain Knowledge Guided Bayesian Optimization For Autonomous Alignment Of Complex Scientific Instruments

Bayesian Optimization (BO) is a powerful tool for optimizing complex non-linear systems. However, its performance degrades in high-dimensional problems with tightly coupled parameters and highly asymmetric objective landscapes, where rewards are sparse. In such needle-in-a-haystack scenarios, even advanced methods like trust-region BO (TurBO) often lead to unsatisfactory results. We propose a domain knowledge guided Bayesian Optimization approach, which leverages physical insight to fundamentally simplify the search problem by transforming coordinates to decouple input features and align the active subspaces with the primary search axes. We demonstrate this approach's efficacy on a challenging 12-dimensional, 6-crystal Split-and-Delay optical system, where conventional approaches, including standard BO, TuRBO and multi-objective BO, consistently led to unsatisfactory results. When combined with an reverse annealing exploration strategy, this approach reliably converges to the global optimum. The coordinate transformation itself is the key to this success, significantly accelerating the search by aligning input co-ordinate axes with the problem's active subspaces. As increasingly complex scientific instruments, from large telescopes to new spectrometers at X-ray Free Electron Lasers are deployed, the demand for robust high-dimensional optimization grows. Our results demonstrate a generalizable paradigm: leveraging physical insight to transform high-dimensional, coupled optimization problems into simpler representations can enable rapid and robust automated tuning for consistent high performance while still retaining current optimization algorithms.

FOS: Computer and information sciences↗

Comparative Performance Evaluation of High-performance Data Transfer Tools

Data transfer in wide-area networks has been long studied in different contexts, from data sharing among data centers to online access to scientific data. Many software tools and platforms have been developed to facilitate easy, reliable, fast, and secure data transfer over wide area networks, such as GridFTP, FDT, bbcp, mdtmFTP, and XDD. However, few studies have shown the full capabilities of existing data transfer tools from the perspective of whether such tools have fully adopted state-of-the-art techniques through meticulous comparative evaluations. In this paper, we evaluate the performance of the four highperformance data transfer tools (GridFTP, FDT, mdtmFTP, and XDD) in various environments. Our evaluation suggests that each tool has strengths and weaknesses. FDT and GridFTP perform consistently in diverse environments. XDD and mdtmFTP show improved performance in limited environments and datasets during our evaluation. Unlike other studies on data transfer tools, we also evaluate the predictability of the tools' performance, an important factor for scheduling different stages of science workflows. Performance predictability also helps in (auto)tuning the configurable parameters of the data transfer tool. We apply statistical learning techniques such as linear/polynomial regression, and k-nearest neighbors (kNN), to assess the performance predictability of each tool using its control parameters. Our results show that we can achieve good prediction performance for GridFTP and mdtmFTP using linear regression and kNN, respectively.

Nadig, Deepak↗

Cu-Local Structures and Their Relation with Nanoscale Electrical Performance in CdTe

For decades, copper has been introduced in CdTe devices to improve overall performance (open circuit voltage, fill factor, and series resistance). While multiple articles have reported on Cu-based defects, very little is known about how the local structure around the copper atom affects electrical performance. Using X-ray Absorption Near Edge Structure (XANES) coupled with X-ray microscopy we investigate good and poor performing region in Cu-doped CdTe devices. Our XANES coupled with theoretical standards by FEFF9 suggest that CU 2 Te phase and Cu Cd may be responsible for the high electrical performance of the regions under study. This correlation of structure-performance at the nanoscale offers a unique framework to understand and tune processes with deep implications to the overall electrical performance of the solar cell.

CdTe↗

Estimating PV power from aggregate power measurements within the distribution grid

The increased integration of photovoltaic (PV) systems in distribution grids reduces visibility and situational awareness for utilities because the PV systems' power production is usually not monitored by them. To address this problem, a method called Contextually Supervised Source Separation (CSSS) has been recently adapted for real-time estimation of aggregate PV active power generation from aggregate net active and reactive power measurements at a point in a radially configured distribution grid (e.g., substation). In its original version, PV disaggregation is formulated as an optimization problem that fits linear regression models for the aggregate PV active power generation and true substation active power load. This paper extends the previous work by adding regularization terms in the objective function to capture additional contextual information such as smoothness, by adding new constraints, by introducing new regressors such as ambient temperature, and by investigating the use of time-varying regressors. Furthermore, we perform extensive parametric analysis to study tuning of the objective function weighting factors in a way that maximizes performance and robustness. The proposed PV disaggregation method can be applied to networks with either a single PV system (e.g., megawatt scale) or many distributed ones (e.g., residential scale) connected downstream of the substation. Simulation studies with real field recorded data show that the enhancements of the proposed method reduce disaggregation error by 58% in winter and 35% in summer compared to previous CSSS-based work. When compared against a commonly used transposition model based approach, the reduction in disaggregation error is more pronounced (78% reduction in winter and 45% in summer). Additional simulations indicate that the proposed algorithm is also applicable for PV systems with time-varying power factors. In conclusion, overall, our results show that—with appropriate modeling and tuning—it is possible to accurately estimate the aggregated PV active power generation of a distribution feeder with minimal or no additional sensor deployment.

14 SOLAR ENERGY↗

Performance Exploration Through Optimistic Static Program Annotations

Compilers are limited by the static information directly or indirectly encoded in the program. Low-level languages, such as C/C++, are considered problematic as their weak type system and relaxed memory semantic allows for various, sometimes non-obvious, behaviors. Since compilers have to preserve the program semantics for all program executions, the existence of exceptional behavior can prevent optimizations that the developer would consider valid and might expect. Analyses to guarantee the absence of disruptive and unlikely situations are consequently an indispensable part of an optimizing compiler. However, such analyses have to be approximative and limited in scope as global and exact solutions are infeasible for any non-trivial program. In this paper, we present an automated tool to measure the effectmissing static information has on the optimizations applied to a given program. The approach generates an optimistically optimized program version which, compared to the original, defines a performance gap that can be closed by better compiler analyses and selective static program annotations. Our evaluation on six already tuned proxy applications for high-performance codes shows speedups of up to 20.6%. This clearly indicates that static uncertainty limits performance. At the same time, we observed that compilers are often unable to utilize additional static information. Thus, manual annotation of all correct static information is therefore not only error prone but also mostly redundant.

Auto Tuning↗

Global climate impacts of stochastic deep convection parameterization in the NCAR CAM5

In this paper, the stochastic deep convection parameterization of Plant and Craig (PC) is implemented in the Community Atmospheric Model version 5 (CAM5) to incorporate the stochastic processes of convection into the Zhang-McFarlane (ZM) deterministic deep convective scheme. Its impacts on deep convection, shallow convection, large-scale precipitation and associated dynamic and thermodynamic fields are investigated. Results show that with the introduction of the PC stochastic parameterization, deep convection is decreased while shallow convection is enhanced. The decrease in deep convection is mainly caused by the stochastic process and the spatial averaging of input quantities for the PC scheme. More detrained liquid water associated with more shallow convection leads to significant increase in liquid water and ice water paths, which increases large-scale precipitation in tropical regions. Specific humidity, relative humidity, zonal wind in the tropics, and precipitable water are all improved. The simulation of shortwave cloud forcing (SWCF) is also improved. The PC stochastic parameterization decreases the global mean SWCF from -52.25 W/m 2 in the standard CAM5 to -48.86 W/m 2 , close to -47.16 W/m 2 in observations. The improvement in SWCF over the tropics is due to decreased low cloud fraction simulated by the stochastic scheme. Sensitivity tests of tuning parameters are also performed to investigate the sensitivity of simulated climatology to uncertain parameters in the stochastic deep convection scheme.

54 ENVIRONMENTAL SCIENCES↗

Anderson acceleration with approximate calculations: Applications to scientific computing

Here we provide rigorous theoretical bounds for Anderson acceleration (AA) that allow for approximate calculations when applied to solve linear problems. We show that, when the approximate calculations satisfy the provided error bounds, the convergence of AA is maintained while the computational time could be reduced. We also provide computable heuristic quantities, guided by the theoretical error bounds, which can be used to automate the tuning of accuracy while performing approximate calculations. For linear problems, the use of heuristics to monitor the error introduced by approximate calculations, combined with the check on monotonicity of the residual, ensures the convergence of the numerical scheme within a prescribed residual tolerance. Motivated by the theoretical studies, we propose a reduced variant of AA, which consists in projecting the least-squares used to compute the Anderson mixing onto a subspace of reduced dimension. The dimensionality of this subspace adapts dynamically at each iteration as prescribed by the computable heuristic quantities. We numerically show and assess the performance of AA with approximate calculations on: (i) linear deterministic fixed-point iterations arising from the Richardson's scheme to solve linear systems with open-source benchmark matrices with various preconditioners and (ii) non-linear deterministic fixed-point iterations arising from non-linear time-dependent Boltzmann equations.

97 MATHEMATICS AND COMPUTING↗