Resolving the GPU responsiveness dilemma through program transformations
Not provided.
Computer Science↗
SEARCH · Search NASA
Results for “program transformation”
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.
ENERGY STAR® Retail Products Platform (RPP): Conditions and Considerations in Evaluating Market Transformation Programs and Evaluation Guidance for RPP
The purpose of this guide is to provide a resource for state utility regulators, utilities, the evaluation community and regulatory stakeholders on methods to measure energy savings from the ENERGY STAR Retail Products Platform (link is external). The guidelines outlined in this document were developed by evaluation experts.
29 ENERGY PLANNING, POLICY, AND ECONOMY↗
The human body at cellular resolution: the NIH Human Biomolecular Atlas Program
Transformative technologies are enabling the construction of three dimensional (3D) maps of tissues with unprecedented spatial and molecular resolution. Over the next seven years, the NIH Common Fund Human Biomolecular Atlas Program (HuBMAP) aims to develop a widely accessible framework for comprehensively mapping the human body at single-cell resolution by supporting technology development, data acquisition, and detailed spatial mapping. HuBMAP will integrate its efforts with other funding agencies, programs, and the biomedical research community at large towards the shared vision of a comprehensive, accessible 3D molecular and cellular atlas of the human body, in health and various disease settings.
59 BASIC BIOLOGICAL SCIENCES↗
Universal Polarization Transformations: Spatial Programming of Polarization Scattering Matrices Using a Deep Learning‐Designed Diffractive Polarization Transformer
Abstract Controlled synthesis of optical fields having nonuniform polarization distributions presents a challenging task. Here, a universal polarization transformer is demonstrated that can synthesize a large set of arbitrarily‐selected, complex‐valued polarization scattering matrices between the polarization states at different positions within its input and output field‐of‐views (FOVs). This framework comprises 2D arrays of linear polarizers positioned between isotropic diffractive layers, each containing tens of thousands of diffractive features with optimizable transmission coefficients. After its deep learning‐based training, this diffractive polarization transformer can successfully implement N i N o = 10 000 different spatially‐encoded polarization scattering matrices with negligible error, where N i and N o represent the number of pixels in the input and output FOVs, respectively. This universal polarization transformation framework is experimentally validated in the terahertz spectrum by fabricating wire‐grid polarizers and integrating them with 3D‐printed diffractive layers to form a physical polarization transformer. Through this set‐up, an all‐optical polarization permutation operation of spatially‐varying polarization fields is demonstrated, and distinct spatially‐encoded polarization scattering matrices are simultaneously implemented between the input and output FOVs of a compact diffractive processor. This framework opens up new avenues for developing novel devices for universal polarization control and may find applications in, e.g., remote sensing, medical imaging, security, material inspection, and machine vision.
Optical neural networks↗
A Transformational Graduate Program: Mines and NREL Join Forces To Shape the Future of Energy
Developing secure, resilient, and adaptive energy infrastructure that fosters economic growth while reducing environmental impact is among the most pressing challenges—and greatest opportunities—of our time. It will require the next generation of thought leaders to capably and creatively guide this vital energy sector transition. Through a new interdisciplinary graduate program, Colorado School of Mines (Mines) and the National Renewable Energy Laboratory (NREL) are cultivating the rich opportunity landscape needed to develop these future innovators. The Mines/NREL Advanced Energy Systems (AES) degree program prepares researchers at the doctoral level and energy professionals at the master’s level to address the full complexity of tomorrow's infrastructure, economic, and environmental challenges. The combination of a thriving cleantech market in Colorado, world-class resources and researchers at NREL, and rigorous educational and technical programs at Mines create a truly one-of-a-kind opportunity for advancing fields related to energy systems. These links can uniquely equip top-caliber students to become the scientists and engineers at the forefront of our global energy economy transformation.
Advanced Energy Systems↗
Assessment of commercially available energy-efficient room air conditioners including models with low global warming potential (GWP) refrigerants
Improving the energy efficiency of room air conditioners (RACs) while transitioning to low global-warming-potential (GWP) refrigerants will be a critical step toward reducing the energy, peak load, and emissions impacts of RACs while keeping costs low. Previous research quantified the benefits of leapfrogging to high efficiency in tandem with the transition to low-GWP refrigerants for RACs (Shah et al., 2015) and identified opportunities for initial action to coordinate energy efficiency with refrigerant transition in economies constituting about 65% of the global RAC market (Shah et al., 2017). This report describes further research performed to identify the best-performing (i.e., most efficient and low-GWP-refrigerant using) RACs on the market, to support an understanding of the best available technology (BAT). Understanding BAT can help support market-transformation programs for high-efficiency and low-GWP equipment such as minimum energy performance standards (MEPS), labeling, procurement, and incentive programs. We studied RACs available in six economies—China, Europe, India, Japan, South Korea, and the United States—that together account for about 70% of global RAC demand, as well as other emerging economies. The following are our key findings: • Highly efficient RACs using low-GWP refrigerants, e.g., HFC-32 (R-32) and HC-290 (R-290), are commercially available today at prices comparable to similar RACs using high-GWP HCFC-22 (R-22) or HFC-410A (R-410A). • High efficiency is typically a feature of high-end products. However, highly efficient, cost-competitive (less than 1,000 or 1,500 U.S. dollars in retail price, depending on size) RACs are available. • Where R-22 is being phased out, high GWP R-410A still dominates RAC sales in most mature markets except Japan, where R-32 dominates. • In all of the economies studied except Japan, only a few models are energy efficient and use low-GWP refrigerants. For example, in Europe, India, and Indonesia, the highest-efficiency RAC models employ the low-GWP refrigerants R-32 or R-290. • RACs are available in most regions and worldwide that surpass the highest efficiency levels recognized by labeling programs. • Fixed-speed RACs using high-GWP and ozone-depleting R-22 refrigerant still dominate the market in many emerging economies. There is significant scope to improve RAC efficiency and transition to low-GWP refrigerants using commercially available technology and to design market-transformation programs for high-efficiency, low-GWP equipment including standards, labeling, procurement, and incentive programs.
29 ENERGY PLANNING, POLICY, AND ECONOMY↗
Analytical Modeling of Cache Behavior for Affine Programs
Modern optimizing compiler design integrates program transformation strategies aimed at reducing data movement to/from main memory by exploiting the data cache hierarchy. But in reality, the effect of these transformations on the actual cache miss count is ignored in compilers due to the lack of precise compile-time models of misses for hierarchical caches. The current state of practice for cache miss analysis is based on accurate simulation. However, simulation requires time proportional to the dataset/problem size as well as the number of distinct cache configurations of interest to be evaluated. This paper takes a fundamentally different approach, made possible by focusing on polyhedral programs with static control flow: instead of relying on costly simulation, this paper develops the first closed-form solution for exact modeling of misses in a set-associative cache hierarchy. It enabling simulation-less selection of program transformations at compile-time to optimize cache misses. A push-button tool implementing the approach is developed and used for extensive validation of the complete framework.
affine programs, cache modeling, set associativity↗
ROSE
Developed at Lawrence Livermore National Laboratory (LLNL), ROSE is an open source compiler infrastructure to build source-to-source program transformation and analysis tools for large-scale C (C89 to C23), C++ (C++98 to C++23), UPC, Fortran (Fortran4, 66, 77, 95, 2003), OpenMP, Java, Python, and Binary applications. ROSE users range from experienced compiler researchers to library and tool developers who may have minimal compiler experience. ROSE is particularly well suited for building custom tools for static analysis, program optimization, arbitrary program transformation, domain-specific optimizations, complex loop optimizations, performance analysis, and cyber-security. ROSE is: A library (and set of associated tools) to quickly and easily apply compiler techniques to one's code in order to improve application performance and developer productivity. A research and development compiler infrastructure for for writing custom source-to-source translators to perform source code transformations, analysis, and optimizations. Is
Pinnow, NathanT [Lawrence Livermore National Labor↗
Cross-Feature Transfer Learning for Efficient Tensor Program Generation
Tuning tensor program generation involves navigating a vast search space to find optimal program transformations and measurements for a program on the target hardware. The complexity of this process is further amplified by the exponential combinations of transformations, especially in heterogeneous environments. This research addresses these challenges by introducing a novel approach that learns the joint neural network and hardware features space, facilitating knowledge transfer to new, unseen target hardware. A comprehensive analysis is conducted on the existing state-of-the-art dataset, TenSet, including a thorough examination of test split strategies and the proposal of methodologies for dataset pruning. Leveraging an attention-inspired technique, we tailor the tuning of tensor programs to embed both neural network and hardware-specific features. Notably, our approach substantially reduces the dataset size by up to 53% compared to the baseline without compromising Pairwise Comparison Accuracy (PCA). Furthermore, our proposed methodology demonstrates competitive or improved mean inference times with only 25–40% of the baseline tuning time across various networks and target hardware. The attention-based tuner can effectively utilize schedules learned from previous hardware program measurements to optimize tensor program tuning on previously unseen hardware, achieving a top-5 accuracy exceeding 90%. This research introduces a significant advancement in autotuning tensor program generation, addressing the complexities associated with heterogeneous environments and showcasing promising results regarding efficiency and accuracy.
97 MATHEMATICS AND COMPUTING↗
A Framework for the Analysis of Compiler Optimizations
Compilers transform program source code to machine executable code. During this transformation, they perform a number of compiler optimizations to improve the performance of the generated executable code. Importantly, applying those optimizations depends on the source code structure, such as the parallel programming model used to parallelize an algorithm. Often, implementations of the same algorithm with different programming models have vastly different performance because the compiler optimized them differently. We create FAROS, a framework to structure and automate the analysis of compiler optimizations on programs. FAROS automates the building process, execution profiling, and analysis of compiler optimization of programs, through a configuration interface. It outputs compiler optimization reports to show which optimizations applied to which line of source code, leveraging compilation remarks output by the compiler. Also, FAROS supports benchmarking performance of different program versions by collecting execution time results. In this first release of FAROS, we provide a configuration file to analyze compiler optimization differences for sequential vs OpenMP compilation, including 38 programs consisting of HPC proxy/mini/large applications, and NAS and Rodina kernels for analysis.
Georgakoudis, Giorgis↗
AstraAI v1
AstraAI is an open-source, structure-aware AI coding agent designed for large scientific and DOE-HPC codebases such as AMReX-based applications. Unlike general-purpose coding assistants, AstraAI combines retrieval-augmented generation (RAG) with compiler-level Abstract Syntax Tree (AST) analysis to perform precise, scope-constrained code modifications. It identifies exact function spans, enforces locality of edits, and maintains cross-file invariants, enabling deterministic and build-safe transformations in complex C++/GPU environments. AstraAI is intended for developers working on large, evolving HPC frameworks where correctness, reproducibility, and structural integrity are critical. Typical use cases include modifying physics kernels, updating GPU device lambdas, and performing multi-file refactors without breaking compilation or runtime semantics. Compared to conventional LLM-based coding agents - even those with repository access - AstraAI provides structural guarantees rather than free-form text patches. It minimizes unintended diffs, prevents scope drift, preserves formatting and build stability, and reduces structural hallucinations. By integrating compiler tooling directly into the generation loop, AstraAI transforms AI-assisted coding from probabilistic text editing into deterministic, structure-preserving program transformation suitable for mission-critical scientific software.
Natarajan, Mahesh [Lawrence Berkeley National Labo↗
Program Correctness, Verification and Testing for Exascale (Corvette)
The goal of this project is to provide tools to assess the correctness of parallel programs written using hybrid parallelism. There is a dire lack of both theoretical and engineering know-how in the area of finding bugs in hybrid or large scale parallel programs, which our research aims to change. In the project we have demonstrated novel approaches in several areas: 1. Low overhead automated and precise detection of concurrency bugs at scale. 2. Using low overhead bug detection tools to guide speculative program transformations for performance. 3. Techniques to reduce the concurrency required to reproduce a bug using partial program restart/replay. 4. Techniques to provide reproducible execution of floating point programs. 5. Techniques for tuning the floating point precision used in codes.
97 MATHEMATICS AND COMPUTING↗
Challenges and Recommended Policies for Simultaneous Global Implementation of Low-GWP Refrigerants and High Efficiency in Room Air Conditioners
Increasing incomes, electrification, and urbanization—as well as a warming world—are driving up the global stock of air conditioners (ACs), particularly in emerging economies with hot climates. AC energy consumption is expected to increase substantially as the global stock of room ACs rises to 1.5 billion in 2030 and 2.5 billion in 2050. Hence, improving AC energy efficiency will be critical to reducing AC energy, cost (consumer lifecycle cost, electricity generation cost, etc.), peak load, and emissions impacts. The 2016 Kigali Amendment to the Montreal Protocol offers an opportunity to improve AC energy efficiency in tandem with the phasedown of high global warming potential (GWP) hydrofluorocarbon (HFC) refrigerants. Based on the most recent information, a literature review, and interviews with manufacturers and industry experts, we find the main barriers to deploying high-efficiency ACs include concerns about market demand and cost, which could be mitigated by appropriately improved design of market-transformation programs such as standards and labeling, incentive, and procurement programs. The main barriers to the low-GWP refrigerant transition include the need for timely revision of safety standards and associated costs for capacity-building activities allowing safe use of low-GWP refrigerants in ACs. Policy action and the market transformation can be accelerated by advancing the refrigerant transition and efficiency improvements in parallel.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
Trends in best-in-class energy-efficient technologies for room air conditioners
Improving the efficiency of room air conditioners (RACs) could provide significant energy and associated emissions savings, particularly in emerging economies with hot climates where the cooling demand is expected to increase dramatically. To help accelerate efficiency improvements, this study identifies “best-in-class” high-efficiency RAC components and products. The findings show that manufacturers tend to minimize manufacturing costs by using RAC designs that are readily available or standardized to their production, and they share components across various models. High-efficiency RAC models use advanced compressor technologies optimized at a low frequency, large heat exchangers with thermodynamically effective materials and designs, highly efficient direct current fan motors, advanced metering devices, and smart sensors for temperature and humidity control. Recently RAC manufacturers have been improving seasonal efficiency – better reflecting part-load operation – especially via variable-speed (inverter) drives for compressor motors. Recent highest-efficiency RAC models use low global warming potential (GWP) refrigerants, having transitioned from conventional high-GWP refrigerants, in regions where RACs that use these low-GWP refrigerants are commercially available. Recently demonstrated innovative technologies show trends toward smart hybrid designs, evaporative cooling, and solid-state materials beyond the conventional vapor-compression technology. This information could help policymakers improve their RAC market-transformation programs to align with the most-efficient global technology.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
Opportunities for Simultaneous Efficiency Improvement and Refrigerant Transition in Air Conditioning
In October of 2016, nearly 200 Parties agreed to amend the Montreal Protocol in Kigali, Rwanda, to phase-down consumption and production of hydrofluorocarbons (HFCs) by 2050. Growth in the use of HFCs, including those currently used as refrigerants in air-conditioning systems, is being driven by demand from emerging economies, hot climates, and rising incomes that are also undergoing rapid urbanization and electrification. Air conditioners (ACs), as an energy-intensive end-use technology, are also covered by a growing number of energy efficiency standards, labeling, procurement, incentive, and other supporting efficiency programs. Therefore, improving room AC energy efficiency and transitioning to low-global warming potential (GWP) refrigerants simultaneously presents significant opportunities to deploy energy efficient technology and reduce the energy and emissions impacts of room ACs, while keeping costs low for consumers. This report aims to provide an initial sense of the opportunities to improve efficiency and transition to low-GWP refrigerants by reviewing the HCFC and HFC regulatory framework and energy efficiency standards and labeling programs in 19 economies that account for roughly 65 percent of global room AC demand. Based on this analysis, we identified key opportunities for coordinated action on efficiency improvement and refrigerant transition for the domestic room AC sector (i.e., ductless mini-split ACs), such as: - implementing or revising standards and labeling programs to improve efficiency levels with the possibility of adding a low-GWP criterion. - combining fixed speed and inverter AC product categories to account for seasonal variations in climate and part-load operating conditions, and using seasonal energy efficiency ratio (SEER) instead of energy efficiency ratio (EER) to better reflect performance of inverter ACs. - implementing market transformation programs, such as bulk procurement programs, to drive down the costs of efficient technology through economies of scale. - aligning timelines for implementing efficiency standards with timelines for refrigerant management plans to coordinate policy actions. - maximizing the energy efficiency improvements of Montreal Protocol investments by coordinating efforts to help keep costs low for consumers and manufacturers during equipment redesign and manufacturing line retooling for refrigerant transition. In order to realize the significant peak load, energy saving and climate benefits of these opportunities associated with improving energy efficiency in tandem with the refrigerant transition, some risks may need to be mitigated, principally safety concerns over risks of accidental ignition with refrigerants rated flammable. This can be mitigated by continuing and accelerating the development of safety standards currently being updated, for example, by the International Electrotechnical Commission (IEC) and improving training for production, installation and maintenance during the use of flammable refrigerants. While some low-GWP refrigerants may cost more than the current baseline refrigerants, the cost of refrigerants is only ~1% of lifecycle costs for an AC. Costs can also be reduced through manufacturing advances and efficiency improvements, particularly if they are supported by policies that encourage technological development through deployment of superefficient ACs at scale. Finally, during the transition, there is a risk of obsolete technology being deployed in markets that either have not updated their standards or have later compliance dates. This risk can be mitigated by updating standards and reviewing them periodically to ensure their effectiveness.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
International Review of Standards and Labeling Programs for Distribution Transformers
Transmission and distribution (T&D) losses in electricity networks represent 8.5% of final energy consumption in the world. In Latin America, T&D losses range between 6% and 20% of final energy consumption, and represent 7% in Chile. Because approximately one-third of T&D losses take place in distribution transformers alone, there is significant potential to save energy and reduce costs and carbon emissions through policy intervention to increase distribution transformer efficiency. A large number of economies around the world have recognized the significant impact of addressing distribution losses and have implemented policies to support market transformation towards more efficient distribution transformers. As a result, there is considerable international experience to be shared and leveraged to inform countries interested in reducing distribution losses through policy intervention. The report builds upon past international studies of standards and labeling (S&L) programs for distribution transformers to present the current energy efficiency programs for distribution transformers around the world.
29 ENERGY PLANNING, POLICY, AND ECONOMY↗
Fast Transient Optimization of Gas Pipelines by Analytic Transformation to Linear Programs
Explore the source record for details and available documents.
Energy Sciences↗
Carbon Storage Program Data Curation, Transformation, and Reuse
For presentation at the Carbon Management Project Review Meeting, Pittsburgh, PA, August 28--September 1, 2023.
Morkner, Paige↗