Search NASA⌕ Search

SEARCH · Search NASA

Results for “Traditional Machine Learning”

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.

1,199 records · Page 67

Characterizing Families of Spectral Similarity Scores and Their Use Cases for Gas Chromatography–Mass Spectrometry Small Molecule Identification

Metabolomics provides a unique snapshot into the world of small molecules and the complex biological processes that govern the human, animal, plant, and environmental ecosystems encapsulated by the One Health modeling framework. However, this “molecular snapshot” is only as informative as the number of metabolites confidently identified within it. The spectral similarity (SS) score is traditionally used to identify compound(s) in mass spectrometry approaches to metabolomics, where spectra are matched to reference libraries of candidate spectra. Unfortunately, there is little consensus on which of the dozens of available SS metrics should be used. This lack of standard SS score creates analytic uncertainty and potentially leads to issues in reproducibility, especially as these data are integrated across other domains. In this work, we use metabolomic spectral similarity as a case study to showcase the challenges in consistency within just one piece of the One Health framework that must be addressed to enable data science approaches for One Health problems. Here, using a large cohort of datasets comprising both standard and complex datasets with expert-verified truth annotations, we evaluated the effectiveness of 66 similarity metrics to delineate between correct matches (true positives) and incorrect matches (true negatives). We additionally characterize the families of these metrics to make informed recommendations for their use. Our results indicate that specific families of metrics (the Inner Product, Correlative, and Intersection families of scores) tend to perform better than others, with no single similarity metric performing optimally for all queried spectra. This work and its findings provide an empirically-based resource for researchers to use in their selection of similarity metrics for GC-MS identification, increasing scientific reproducibility through taking steps towards standardizing identification workflows.

59 BASIC BIOLOGICAL SCIENCES↗

Advances in Hyperspectral Image Classification Methods for Vegetation and Agricultural Cropland Studies

Hyperspectral data are becoming more widely available via sensors on airborne and unmanned aerial vehicle (UAV) platforms, as well as proximal platforms. While space-based hyperspectral data continue to be limited in availability, multiple spaceborne Earth-observing missions on traditional platforms are scheduled for launch, and companies are experimenting with small satellites for constellations to observe the Earth, as well as for planetary missions. Land cover mapping via classification is one of the most important applications of hyperspectral remote sensing and will increase in significance as time series of imagery are more readily available. However, while the narrow bands of hyperspectral data provide new opportunities for chemistry-based modeling and mapping, challenges remain. Hyperspectral data are high dimensional, and many bands are highly correlated or irrelevant for a given classification problem. For supervised classification methods, the quantity of training data is typically limited relative to the dimension of the input space. The resulting Hughes phenomenon, often referred to as the curse of dimensionality, increases potential for unstable parameter estimates, overfitting, and poor generalization of classifiers. This is particularly problematic for parametric approaches such as Gaussian maximum likelihood–based classifiers that have been the backbone of pixel-based multispectral classification methods. This issue has motivated investigation of alternatives, including regularization of the class covariance matrices, ensembles of weak classifiers, development of feature selection and extraction methods, adoption of nonparametric classifiers, and exploration of methods to exploit unlabeled samples via semi-supervised and active learning. Data sets are also quite large, motivating computationally efficient algorithms and implementations. This chapter provides an overview of the recent advances in classification methods for mapping vegetation using hyperspectral data. Three data sets that are used in the hyperspectral classification literature (e.g., Botswana Hyperion satellite data and AVIRIS airborne data over both Kennedy Space Center and Indian Pines) are described in Section 3.2 and used to illustrate methods described in the chapter. An additional high-resolution hyperspectral data set acquired by a SpecTIR sensor on an airborne platform over the Indian Pines area is included to exemplify the use of new deep learning approaches, and a multiplatform example of airborne hyperspectral data is provided to demonstrate transfer learning in hyperspectral image classification. Classical approaches for supervised and unsupervised feature selection and extraction are reviewed in Section 3.3. In particular, nonlinearities exhibited in hyperspectral imagery have motivated development of nonlinear feature extraction methods in manifold learning, which are outlined in Section 3.3.1.4. Spatial context is also important in classification of both natural vegetation with complex textural patterns and large agricultural fields with significant local variability within fields. Approaches to exploit spatial features at both the pixel level (e.g., co-occurrence–based texture and extended morphological attribute profiles [EMAPs]) and integration of segmentation approaches (e.g., HSeg) are discussed in this context in Section 3.3.2. Recently, classification methods that leverage nonparametric methods originating in the machine learning community have grown in popularity. An overview of both widely used and newly emerging approaches, including support vector machines (SVMs), Gaussian mixture models, and deep learning based on convolutional neural networks is provided in Section 3.4. Strategies to exploit unlabeled samples, including active learning and metric learning, which combine feature extraction and augmentation of the pool of training samples in an active learning framework, are outlined in Section 3.5. Integration of image segmentation with classification to accommodate spatial coherence typically observed in vegetation is also explored, including as an integrated active learning system. Exploitation of multisensor strategies for augmenting the pool of training samples is investigated via a transfer learning framework in Section 3.5.1.2. Finally, we look to the future, considering opportunities soon to be provided by new paradigms, as hyperspectral sensing is becoming common at multiple scales from ground-based and airborne autonomous vehicles to manned aircraft and space-based platforms.

Pasolli, Edoardo↗

Inverse Reinforcement Learning based Bayesian Goal Inference Method for Early Nuclear Proliferation Detection

Traditional methods for detection of nuclear proliferation indicators are usually applied after nuclear proliferation has already occurred. There is a need to advance these methods to perform early detection of nuclear proliferation indicators. In this project, we formulated an early detection problem as a sequential, decision-making, goal inference problem based on research publications of authors, to determine whether it is possible to infer whether an author will publish on a research activity before it has occurred. To develop and test our approach, we selected a civil nuclear activity for our case study. We constructed a state-action-state transition graph from publications of authors associated with the activity and the co-authors of their publications, using titles, abstracts, and author publication sequences. We then used inverse reinforcement learning to model the goal-directed behavior of authors in trajectories that terminate at selected goal states. Using a Bayesian formulation, we computed the probability that authors would reach each selected state from partially observed trajectories of their state transitions in their research topic space. The state with the highest probability was selected as the most probable goal state. Based on our results, we found that 60% of the times we can infer the correct goal state early; sometimes the inference is either delayed, or multiple states could be inferred as goal states. Overall, our results show that it is possible to perform early detection of research activities of authors in a nuclear technology area. Further research is necessary to establish a more accurate understanding of how topic modeling, topic space grid discretization, and the extent of overlap among trajectories of different goal states, affect the goal inference results. The methods developed in this work may be used to enhance data-driven methods for early detection of nuclear proliferation indicators.

97 MATHEMATICS AND COMPUTING↗

Real World Applications of AI/ML in Optimizing Airspace Operations

As National Airspace System (NAS) is going through the Digital Transformation journey, data science and analytics methods can significantly contribute to improving the traditional physics-based decision-making tools. The adoption of AI/ML methods will not only help accelerate Federal Aviation Administration (FAA)’s vision of Info-centric NAS but also contribute to the overall objective of sustainable aviation. AI/ML can improve the ground and airspace operations by enhancing the accuracy of current decision-making tools used by the airlines and the FAA to manage traffic on the ground and in the air. Huge amount of data that gets collected during a flight. AI/ML methods can extract information from this data and provide valuable insights to make better operational decisions. NASA has partnered with the FAA and commercial airlines such as American and Southwest Airlines on this effort and has successfully demonstrated the benefits of using ML in real world environment by reducing delays and optimizing ground operations at the US airports. In 2022 itself, NASA demonstrated real-world benefits (over 24K lbs. of fuel savings, over 76.6K lbs. CO2 emission savings, and several hours of delay savings) by deploying ML based prediction models to optimize ground operations at Dallas/Fort Worth International and Dallas Love Field Airports in Texas. These tools are being deployed on the cloud for broader deployment, adaptability, and scalability. NASA is developed a reference implementation of the cloud-based platform to significantly lower the bar to development and distribution of these digital services for aviation. In this talk, I will share information about the Digital Information Platform project, the novel AI/ML based approaches used for optimizing ground operations and the opportunities to partner with NASA on these demonstrations.

air traffic management↗

A physics-constrained neural ordinary differential equations approach for robust learning of stiff chemical kinetics

The high computational cost associated with solving for detailed chemistry poses a significant challenge for predictive computational fluid dynamics (CFD) simulations of turbulent reacting flows. While deep learning techniques have been explored to develop faster surrogate models, they often fail to integrate reliably with CFD solvers. This instability arises because traditional deep learning approaches optimize for training error without ensuring compatibility with ordinary differential equation (ODE) solvers, resulting in accumulation of errors over time. Recently, neuralODE (NODE) based approaches have been shown to be a promising technique to emulate and accelerate detailed chemistry computations. Here, in the present work, we extend this NODE framework for stiff chemical kinetics by incorporating mass conservation constraints directly into the loss function during training. This ensures that the total mass as well as the individual elemental species masses are conserved in an a-posteriori manner. Proof-of-concept studies are performed with the novel physics-constrained NODE (PC-NODE) approach for homogeneous autoignition of hydrogen-air mixture over a range of composition and thermodynamic conditions. It is demonstrated that the PC-NODE framework not only improves the physical consistency of the resulting data-driven model with respect to mass conservation criteria, but also improves training efficiency. PC-NODE is shown to achieve 2–100× speedup relative to the hydrogen-air detailed chemical mechanism depending on the type of the ODE solver (implicit or explicit) used during autoregressive inference tests. Lastly, a-posteriori studies are performed wherein the trained PC-NODE model is coupled with a CFD solver. It is shown that higher accuracy is achieved with PC-NODE relative to the purely data-driven NODE approach. Moreover, PC-NODE also exhibits robustness and generalizability to unseen initial conditions from within (interpolative capability) as well as outside (extrapolative capability) the training regime.

computational combustion↗

Data-driven cyber attack detection and mitigation for decentralized wide-area protection and control in smart grids

Modern power systems have already evolved into complicated cyber physical systems (CPS), often referred to as smart grids, due to the continuous expansion of the electrical infrastructure, the augmentation of the number of heterogeneous system components and players, and the consequential application of a diversity of information and telecommunication technologies to facilitate the Wide Area Monitoring, Protection and Control (WAMPAC) of the day-to-day power system operation. Because of the reliance on cyber technologies, WAMPAC, among other critical functions, is prone to various malicious cyber attacks. Successful cyber attacks, especially those sabotage the operation of Bulk Electric System (BES), can cause great financial losses and social panics. Application of conventional IT security solutions is indispensable, but it often turns out to be insufficient to mitigate sophisticated attacks that deploy zero-day vulnerabilities or social engineering tactics. To further improve the resilience of the operation of smart grids when facing cyber attacks, it is desirable to make the WAMPAC functions per se capable of detecting various anomalies automatically, carrying out adaptive activity adjustments in time and thus staying unimpaired even under attack. Most of the existing research efforts attempt to achieve this by adding novel functional modules, such as model-based anomaly detectors, to the legacy centralized WAMPAC functions. In contrast, this dissertation investigates the application of data-driven algorithms in cyber attack detection and mitigation within a decentralized architecture aiming at improving the situational awareness and self-adaptiveness of WAMPAC. First part of the research focuses on the decentralization of System Integrity Protection Scheme (SIPS) with Multi-Agent System (MAS), within which the data-driven anomaly detection and optimal adaptive load shedding are further explored. An algorithm named as Support Vector Machine embedded Layered Decision Tree (SVMLDT) is proposed for the anomaly detection, which provides satisfactory detection accuracy as well as decision-making interpretability. The adaptive load shedding is carried out by every agent individually with dynamic programming. The load shedding relies on the load profile propagation among peer agents and the attack adaptiveness is accomplished by maintaining the historical mean of load shedding proportion. Load shedding only takes place after the consensus pertaining to the anomaly detection is achieved among all interconnected agents and it serves the purpose of mitigating certain cyber attacks. The attack resilience of the decentralized SIPS is evaluated using IEEE 39 bus model. It is shown that, unlike the traditional centralized SIPS, the proposed solution is able to carry out the remedial actions under most Denial of Service (DoS) attacks. The second part investigates the clustering based anomalous behavior detection and peer-assisted mitigation for power system generation control. To reduce the dimensionality of the data, three metrics are designed to interpret the behavior conformity of generator within the same balancing area. Semi-supervised K-means clustering and a density sensitive clustering algorithm based on Hierarchical DBSCAN (HDBSCAN) are both applied in clustering in the 3D feature space. Aiming to mitigate the cyber attacks targeting the generation control commands, a peer-assisted strategy is proposed. When the control commands from control center is detected as anomalous, i.e. either missing or the payload of which have been manipulated, the generating unit utilizes the peer data to infer and estimate a new generation adjustment value as replacement. Linear regression is utilized to obtain the relation of control values received by different generating units, Moving Target Defense (MTD) is adopted during the peer selection and 1-dimensional clustering is performed with the inferred control values, which are followed by the final control value estimation. The mitigation strategy proposed requires that generating units can communicate with each other in a peer-to-peer manner. Evaluation results suggest the efficacy of the proposed solution in counteracting data availability and data integrity attacks targeting the generation controls. However, the strategy stays effective only if less than half of the generating units are compromised and it is not able to mitigate cyber attacks targeting the measurements involved in the generation control.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Artificial neural network based predictions of cetane number for furanic biofuel additives

The next generation of alternative fuels is being explored through advanced chemical and biological production techniques for the purpose of finding suitable replacements for diesel and gasoline while lowering production costs and increasing process yields. Chemical conversion of biomass to fuels offers a plethora of pathways with a variety of fuel molecules, both novel and traditional, which may be targeted. In the search for new fuels, an initial, intuition-driven evaluation of fuel compounds with desired properties is required. Due to the high cost and significant production time needed to synthesize these materials at a scale sufficient for exhaustive testing, a predictive model would allow chemists to preemptively screen fuel properties of potentially desirable fuel candidates. Current work has shown that predictive models, in this case artificial neural networks (ANN's) analyzing quantitative structure property relationships (QSPR's), can predict the cetane number (CN) of a proposed fuel molecule with relatively small error. A fuel's CN is a measure of its ignition quality, typically defined using prescribed ASTM standards and a cetane testing engine. Alternatively, the analogous derived cetane number (DCN), obtained using an Ignition Quality Tester (IQT), is a direct measurement alternative to the CN that uses an empirical inverse relationship to the ignition delay found in the constant volume combustion chamber apparatus. DCN data points acquired using an IQT were utilized for model validation and expansion of the experimental database used in this study. The present report improves on an existing model by optimizing the model architecture along with the key learning variables of the ANN and by making the model more generalizable to a wider variety of fuel candidate types, specifically the class of furans and furan derivatives, by including specific molecules for the model to incorporate. The new molecules considered include tetrahydrofuran, 2-methylfuran, 2-methyltetrahydrofuran, 5,5'-(furan-2-ylmethylene)bis(2-methylfuran), 5,5'-((tetrahydrofuran-2-yl)methylene)bis(2-methyltetrahydrofuran), tris(5-methylfuran-2-yl)methane, and tris(5-methyltetrahydrofuran-2-yl)methane. Model architecture adjustments improved the overall root-mean-square error (RMSE) of the base database predictions by 5.54%. Additionally, through the targeted database expansion, it is shown that the predicted cetane number of the furan-based molecules improves on average by 49.21% (3.74 CN units) and significantly for a few of the individual molecules. This indicates that a selected subset of representative molecules can be used to extend the model's predictive accuracy to new molecular classes. The approach, bolstered by the improvements presented in this paper, enables chemists to focus on promising molecules by eliminating less favorable candidates in relation to their ignition quality.

09 BIOMASS FUELS↗

Multiscale Characterization of Additive Manufacturing Components with Computed Tomography, 3D X-ray Microscopy, and Deep Learning

Additive manufacturing (AM) facilitates the creation of complex-geometry parts, driving advancements in lightweight aerospace components, high-efficiency engine cooling channels, and customized medical implants. However, ensuring the quality and reliability of AM parts remains challenging due to internal defects, surface irregularities, porosity, and residual trapped powder, which are often inaccessible to traditional inspection methods. Recent developments in X-ray computed tomography (XCT) and 3D X-ray microscopy (XRM), particularly systems equipped with resolution-at-a-distance (RaaD™) capabilities, enable high-resolution, non-destructive evaluation of AM components across multiple scales, from sub-micrometer to macroscopic levels. This paper explores modern XCT and XRM techniques for multiscale characterization of AM parts, focusing on their ability to detect and analyze defects such as porosity, cracks, inclusions, and surface roughness, while offering insights into defect formation mechanisms, material properties, and process-induced variations. The integration of deep learning (DL) frameworks, including Simurgh, DeepRecon, and DeepScout, enhances XCT/XRM workflows by reducing scan times, improving resolution recovery, and enabling accurate defect detection even with limited projection data. These DL-based methods overcome limitations of traditional reconstruction techniques, enabling faster, more reliable characterization of dense materials like Inconel 718 and novel alloys such as AlCe. Applications include process parameter optimization, high-throughput quality control, and multistage AM process evaluation, with DL-enhanced workflows accelerating analysis times from weeks to days. Correlative imaging approaches further validate XCT and XRM data against scanning electron microscopy (SEM) images of physically sectioned samples, confirming the accuracy of DL-based reconstructions and enabling comprehensive defect analysis. While challenges remain in generalizing DL models to diverse materials and imaging conditions, improvements in resolution, noise reduction, and defect detection highlight the transformative potential of these methods. This multiscale and correlative approach enables precise identification and correlation of microstructural features with the overall performance of AM components. By integrating advanced XCT, XRM, and DL techniques, this paper demonstrates a significant leap forward in AM characterization, offering valuable insights into the relationships between processing parameters, microstructure, and part performance, and driving innovations that enhance the quality and reliability of AM products for demanding industrial applications.

Additive manufacturing↗

Subtask 1.4 - Techno-Economic Assessment of Regional Carbon Utilization Scenarios and Attendant Monitoring Technology

The U.S. Department of Energy (DOE) Carbon Storage Program seeks to develop technologies to significantly improve the effectiveness of carbon capture and storage (CCS), reduce the cost of implementation, and support the readiness of widespread commercial deployment by 2035. To accomplish this goal, technical and economic barriers must be addressed and the results used to inform regulators and industry on the safety, permanence, and viability of carbon storage. To help overcome these barriers, the Energy & Environmental Research Center (EERC) conducted three activities: 1) investigate the regional techno-economic life cycle of CO2 enhanced oil recovery (EOR) in North Dakota, 2) assist DOE in determining the prospective CO2 storage resource of residual oil zones (ROZs), and 3) develop a seismic technique for monitoring geologic storage of CO2. These three activities address two strategic goals enumerated in the DOE–EERC Cooperative Agreement, namely, 1) to advance foundational science and 2) to inform data-driven policies that enhance U.S. economic growth and job creation, energy security, and environmental quality. Activity 1 In North Dakota, there is synergistic incentive to develop and implement CO2 EOR technologies. Regional industry partners are actively seeking options that can cost-effectively improve the efficiency of power production, while also coproducing CO2 for use in EOR applications. A full-value-chain techno-economic assessment that links lignite resources, power generation, and EOR is needed to determine the implications for a state or region to continue to use these resources or look elsewhere for other energy sources and revenue streams. The integration of North Dakota’s lignite power industry with the state’s oil industry has the potential to revitalize legacy oil fields and produce up to 1 billion barrels of additional oil using CO2 EOR while providing a market for a CO2 commodity captured at coal-fired power plants. Revitalization of North Dakota’s legacy oil fields will bring new industry investment to the state, resulting in the growth of oil-related jobs and revenue through oil sales, severance tax, and associated oilfield development activities (drilling, pipelines, etc.). CO2 demand for EOR and incremental oil production were calculated for 201 conventional oil fields in North Dakota. Infrastructure development in the fields was estimated using 1) a traditional vertical well arrangement of CO2 injectors and oil producers and 2) an advanced case where horizontal wells were used exclusively. Pipeline, compression, and recycling infrastructure requirements were also estimated. When viewed from the CO2 capture and CO2 EOR sectors, the advanced case showed a positive investment scenario when including potential benefits of the 45Q tax credit policy. An economic evaluation of a full implementation of CO2 EOR showed an average $60 billion potential state economic impact estimated over the 37-year construction and production time frame could be realized with an assumed oil price of $65/bbl. Oil price ranges of $35–$90/bbl cause an estimated variation of ±$5 billion over the potential effort. The total net state government revenues were estimated over the 37-year construction and production time frame. With an assumed oil price of $65/bbl, total potential fiscal impact could average $7.4 billion. This result ranges about ±$2.5 billion for a potential EOR implementation effort for oil price ranges of $35–$90/bbl. This broad-brush assessment was focused on “magnitude-of-effect” perspectives of what CO2 capture and use for EOR could mean to the state. Whether CO2 EOR ever becomes widely adopted is not the issue at this stage of evaluation. The question being answered is, How valuable could CO2 capture and CO2 EOR be for the state and local jurisdictions in western North Dakota? The results of the current economic assessment remove any ambiguity to that question. Full implementation of CO2 EOR as modeled in this study provides substantial economic benefits to the state, even at low oil prices (i.e., $35/bbl). Job creation and government revenues could be substantial, depending on the scale of implementation. With a better understanding of what CO2 EOR might mean for the state’s economy, the focus can now shift to the policies and financial mechanisms needed to make private investment in CO2 capture and use as EOR competitive with other energy-related pursuits. Activity 2 ROZs contain remnants of oil not swept away by natural waterflooding. Although nearly all of the identified ROZs are found in West Texas, there is evidence that this category of oil reservoir can be found in other major oil-producing basins of the United States. Because the formation of these oil reservoirs is distinct from other conventional reservoirs, the method to estimate the prospective CO2 storage resource in ROZs may be equally distinct from the method used for conventional fields. As such, the EERC collaborated with DOE’s National Energy Technology Laboratory (NETL) to investigate the feasibility of CO2 storage in ROZs and determine what specific modification (if any) needs to be made to the current DOE volumetric method to estimate prospective CO2 storage resource in oil reservoirs. This activity built on a previous EERC investigation to identify ROZs in the Williston and Powder River Basins and upon collaborations between the EERC and NETL to develop storage resource estimation methods for different types of geologic CO2 storage. Review of the volumetric approach raised questions about the applicability of such an approach to hydrocarbon-bearing zones without the inclusion of potential hydrocarbon production. A production-based approach was presented by the EERC to inform the working group of additional methods of increasing the CO2 storage resource potential. Because the production-based method focuses more heavily on oil recovery than CO2 storage, it was not included in the working group manuscript. Activity 3 The EERC has used sparse and irregular seismic arrays to detect CO2 injected into a reservoir. The variability of seismic noise recorded by the sensors directly impacts the seismic interpretation. However, designing noise attenuation workflows effectively can be a more challenging task using sparse and irregular arrays for data acquisition. Based on the hypothesis that noncoherent noise is stationary during weekly data acquisition, a new machine-learning algorithm was developed to automatically estimate noise conditions using passive records, noise from active records, and ancillary information. The new noise attenuation processing workflow can improve the time-lapse detection of CO2 from sparse and irregular array acquisitions. Results show spatial and temporal noise patterns can be related to the location of the sensors and the weather conditions during the data acquisition and that our attenuation processing workflow can improve the time-lapse detection of CO2 from sparse and irregular array acquisitions. This effort was funded in part through the EERC–DOE Joint Program on Research and Development for Fossil Energy-Related Resources Cooperative Agreement No. DE-FE0024233. Nonfederal funding was provided by the North Dakota Department of Commerce.

Peck, Wesly↗

Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)

The development of modern processors exhibits two trends that complicate the optimizations of modern software. The first is the increasing sensitivity of processors' throughput to irregularities in computation. With more processors produced through a massive integration of simple cores, future systems will increasingly favor regular data-level parallel computations, but deviate from the needs of applications with complex patterns. Some evidences are already shown on Graphic Processing Units (GPU): Irregular data accesses (e.g., indirect references A[D[i]]) and conditional branches are limiting many GPU applications' performance at a level an order of magnitude lower than the peak of GPU. The second hardware trend is the growing gap between memory bandwidth and the aggregate speed—that is, the sum of all cores' computing power—of a Chip Multiprocessor (CMP). Despite the capped growth of the peak CPU speed, the aggregate speed of a CMP keeps increasing as more cores get into a single chip. It is expected that by 2018, node concurrency in an exascale system will increase by hundreds of times, whereas, memory bandwidth will expand by only 10 to 20 times. Consequently, data movement and storage is expected to consume more than 70% of the total system power. Bridging this gap is difficult; the complexities of modern CMP memory hierarchy make it even harder: Data cache becomes shared among computing units, and the sharing is often non-uniform—whether two computing units share a cache depends on their proximity and the level of the cache. On the recent IBM Power7 architecture, for instance, four hardware contexts (or SMT threads) in a core share the entire memory hierarchy, all cores in one chip share an on-chip L3 cache, and cores across chips share L3 and main memory through off-chip connections. These two trends complicate the translation of computing power into performance, especially for a program with either intensive data accesses or complex patterns in data accesses or control flow paths. Unfortunately, both attributes present and will persist in a class of important applications. For instance, many scientific simulations deal with a large volume of data. And meanwhile, as most real-world processes are non-uniform and evolving (e.g., the evolution of a galaxy or the process of a drug injection), both the computations and data accesses of these programs tend to be irregular and dynamically changing. Currently, the lack of support to these applications on modern CMP severely limits their performance. On GPU, as our recent study shows and other studies echo, performance enhancement of a factor of integers is possible when memory accesses or control flows are streamlined for a set of GPU applications. On multicore CPU, our studies show that traditional locality enhancement, for being oblivious to the new features of multicore memory hierarchy, may even cause large slowdown to data-intensive dynamic applications. The severity of the issues is expected to worsen as the two hardware trends continue. Some recent studies try to match software with the trends, but in a limited scope or manner. For irregularities on GPU, most studies focus on irregularities analyzable through static analysis (e.g., data accesses in regular loops). Dynamic irregularities are harder to address because the needed analysis and transformations typically have to happen at run time. Some other research resorts to hardware extensions, an actual adoption of which is unclear for the entailed space cost and complexity. For data locality, recent years have seen some exploitations of the new memory hierarchy on multicore for performance, but most of them are on process or thread scheduling, rather than program transformations. Our recent study reveals that program-level transformations may magnify the scheduling benefits by a factor of seven, concluding that program-level transformation should play a central role for data locality enhancement on modern CMP. But research in this direction has been sparse, and most have focused on data layout or cache performance modeling, rather than program transformations to match with the new memory hierarchy features. Overall, it is still an open question how to bridge the gap between dynamic computations and the two prominent properties of modern processors. The goal of this project is to develop a set of techniques and software tools to enhance the matching between memory accesses in dynamic simulations and the prominent features of modern and future CMP, alleviating the memory performance issues for petascale and exascale computing. This report summarizes the discoveries and products produced throughout this project. It includes free launch, a new software approach to overcoming the shortcomings of both methods; coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution; algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Supporting dynamic parallelism is important for GPU to benefit a broad range of applications. There are currently two fundamental ways for programs to exploit dynamic parallelism on GPU: a software-based approach with software-managed worklists, and a hardware-based approach through dynamic subkernel launches. Neither is satisfactory. The former is complicated to program and is often subject to some load imbalance; the latter suffers large runtime overhead. In this work, we propose free launch, a new software approach to overcoming the shortcomings of both methods. It allows programmers to use subkernel launches to express dynamic parallelism. It employs a novel compiler-based code transformation named subkernel launch removal to replace the subkernel launches with the reuse of parent threads. Coupled with an adaptive task assignment mechanism, the transformation reassigns the tasks in the subkernels to the parent threads with a good load balance. The technique requires no hardware extensions, immediately deployable on existing GPUs. It keeps the programming convenience of the subkernel launch-based approach while avoiding its large runtime overhead. Meanwhile, its superior load balancing makes it outperform manual worklist-based techniques by 3X on average. The work was published at the 48th Annual IEEE/ACM International Symposium on Microarchitecture (Micro'2015). A Graphic Processing Unit (GPU) system is typically equipped with many types of memory (e.g., global, constant, texture, shared, cache). Data placement determines what data are placed on which type of memory, essential for GPU memory performance. Prior optimizations of data placement always require a single view of a data object on memory, which limits the optimization effectiveness. In this work, we propose coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution. We demonstrate that under certain conditions, the multiple views can remain incoherent while facilitating enhanced data placement. We present a theorem and some compiler support to ensure the soundness of the usage of coherence-free multiview. We further develop reference-discerning data placement, a new way to enhance data placements on GPU. It enables more flexible data placements by using coherence-free multiview to leverage the slack in coherence requirement of some GPU programs. Experiments on three types of GPU systems show that, with less than 200KB space cost, the new data placement technique can provide a 1.6X average (up to 4.27X) speedup. In addition, we have examined algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Computing distances among data points is an essential part of many important algorithms in data analytics, graph analysis, and other domains. In each of these domains, developers have spent significant manual effort optimizing algorithms, often through novel applications of the triangle equality, in order to minimize the number of distance computations in the algorithms. In this work, we observe that many algorithms across these domains can be generalized as an instance of a generic distance-related abstraction. Based on this abstraction, we derive seven principles for correctly applying the triangular inequality to optimize distance-related algorithms. Guided by the findings, we develop Triangular OPtimizer (TOP), the first software framework that is able to automatically produce optimized algorithms that either matches or outperforms manually designed algorithms for solving distance-related problems. TOP achieves up to 237x speedups and 2.5X on average. The work has been published at the 32nd International Conference on Machine Learning (ICML'15) and the 41st International Conference on Very Large Data Bases (VLDB'15).

42 ENGINEERING↗

Innovating the next generation of commercial smart building software

Nearly 30% of commercial building energy use is wasted due to equipment faults and HVAC controls problems. The result is increased emissions, compromised comfort and productivity, and less reliable coordination of building power needs with a clean grid. The energy impact alone represents $17 billion in potential savings. Today’s smart building software provides a robust solution to address these operational deficiencies. Energy management and information systems (EMIS) are saving up to 9% on average, with two-year paybacks. They are being incorporated into energy management processes, commissioning services, and utility programs. As effective as they are, two barriers prevent even deeper benefits; limited personnel to fix problems once they are identified, and the expense and time to manually implement changes in control systems. In partnership with the research community, the EMIS industry is developing new capabilities to overcome these barriers. Moving beyond siloed products for either fault detection and diagnostics, or optimal control, these new capabilities empower users to not only automatically identify faults, but also to push corrective action, and control improvements to their buildings. In this paper, several areas for enhancements are documented: ‘one-time’ correction of faults such as setpoints, schedules, and economizer lockouts; short-term active testing for automated proportional integral derivative (PID) loop tuning and functional testing; and continuous supervisory control for demand flexibility and year-round efficiency. Results are presented from a pair of partner implementations out of a dozen providers integrating these enhancements into their products, including field tests from across the country, and insights into operator acceptance and integration into operations and maintenance practices.

Casillas, Armando↗