Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Graph properties”

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.

569 records · Page 32
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Search for Stable and Low-Energy Ce–Co–Cu Ternary Compounds Using Machine Learning

Cerium-based intermetallics have garnered significant research attention as potential new permanent magnets. In this study, we explore the compositional and structural landscape of Ce−Co−Cu ternary compounds using a machine learning (ML)- guided framework integrated with first-principles calculations. We employ a crystal graph convolutional neural network (CGCNN), which enables efficient screening for promising candidates, significantly accelerating the material discovery process. With this approach, we predict five stable compounds, Ce 3 Co 3 Cu, CeCoCu 2 , Ce 12 Co 7 Cu, Ce 11 Co 9 Cu, and Ce 10 Co 11 Cu 4 , with formation energies below the convex hull, along with hundreds of low-energy (possibly metastable) Ce−Co−Cu ternary compounds. Firstprinciples calculations reveal that several structures are both energetically and dynamically stable. Notably, two Co-rich low-energy compounds, Ce 4 Co 33 Cu and Ce 4 Co 31 Cu 3 , are predicted to have high magnetizations.

Chemical structure↗
NASA NTRS2004

Optical Calibration Process Developed for Neural-Network-Based Optical Nondestructive Evaluation Method

A completely optical calibration process has been developed at Glenn for calibrating a neural-network-based nondestructive evaluation (NDE) method. The NDE method itself detects very small changes in the characteristic patterns or vibration mode shapes of vibrating structures as discussed in many references. The mode shapes or characteristic patterns are recorded using television or electronic holography and change when a structure experiences, for example, cracking, debonds, or variations in fastener properties. An artificial neural network can be trained to be very sensitive to changes in the mode shapes, but quantifying or calibrating that sensitivity in a consistent, meaningful, and deliverable manner has been challenging. The standard calibration approach has been difficult to implement, where the response to damage of the trained neural network is compared with the responses of vibration-measurement sensors. In particular, the vibration-measurement sensors are intrusive, insufficiently sensitive, and not numerous enough. In response to these difficulties, a completely optical alternative to the standard calibration approach was proposed and tested successfully. Specifically, the vibration mode to be monitored for structural damage was intentionally contaminated with known amounts of another mode, and the response of the trained neural network was measured as a function of the peak-to-peak amplitude of the contaminating mode. The neural network calibration technique essentially uses the vibration mode shapes of the undamaged structure as standards against which the changed mode shapes are compared. The published response of the network can be made nearly independent of the contaminating mode, if enough vibration modes are used to train the net. The sensitivity of the neural network can be adjusted for the environment in which the test is to be conducted. The response of a neural network trained with measured vibration patterns for use on a vibration isolation table in the presence of various sources of laboratory noise is shown. The output of the neural network is called the degradable classification index. The curve was generated by a simultaneous comparison of means, and it shows a peak-to-peak sensitivity of about 100 nm. The following graph uses model generated data from a compressor blade to show that much higher sensitivities are possible when the environment can be controlled better. The peak-to-peak sensitivity here is about 20 nm. The training procedure was modified for the second graph, and the data were subjected to an intensity-dependent transformation called folding. All the measurements for this approach to calibration were optical. The peak-to-peak amplitudes of the vibration modes were measured using heterodyne interferometry, and the modes themselves were recorded using television (electronic) holography.

Decker, Arthur J.↗
DOE OSTI2025

A quantitative comparison of the fingerprint of twinned microstructures through surface and three-dimensional techniques

Assessing the fingerprint of a material’s microstructure is key for supporting materials design. With the emergence of a wide range of 3D characterization techniques, it is critical to understand the main differences in fingerprints reconstructed from 2D and 3D datasets. To this end, we introduce a graph-based microstructure reconstruction framework that enables structural comparisons of twin domain networks in high purity Ti using 3D and 2D electron backscatter diffraction. Insights into the structure of the twin networks are facilitated by combining statistical analysis of twin crystallography with visual and graphical analysis of the novel graph abstractions of the twins. We demonstrate that compared to 3D reconstructions, conventional 2D views of twinning miss key aspects of the microstructure including the high interconnectivity of domains into networks that span the full reconstruction volume. The reduced cross-grain and in-grain twin connectivity typically observed in 2D has notable implications on our understanding of how twinning mediates the plastic response of microstructures and how twin networks evolve. It is thus clear that 3D characterization is critical for accurately inferring both twin network morphologies as well as the key unit processes facilitating network formation.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Verification Testing of OLI Systems Mixed Solvent Electrolyte Model for the Na-K-Mg-Ca-H-Cl-SO 4 -OH-HCO 3 -CO 3 -CO 2 -H 2 ) System to High Ionic Strength at 25°C

This technical report summarizes model verification results and summary statistics for 41 evaporite mineral solubility cases evaluated by Savannah River National Laboratory using OLI Systems’ aqueous electrolyte thermodynamic modeling software. The 41 verification cases containing a total of 60 solubility curves comprise mineral solubility data from low to high ionic strength at 25°C for the eight-component system Na-K-Mg-Ca-H-Cl-SO 4 -OH-HCO 3 -CO 3 -CO 2 -H 2 O as reported by Harvie et al. (1984). Thermodynamic calculations were executed using OLI Systems’ Stream Analyzer computation module within the OLI Studio software platform (Ver. 11.0, Rev. 11.0.1.9). The Mixed Solvent Electrolyte (MSE) thermodynamic framework was chosen for this investigation because of its superiority in modeling high ionic-strength inorganic salt solutions and actinide redox chemistry and solubility, both of which are relevant to the geological repository conditions at the Waste Isolation Pilot Plant in Carlsbad, New Mexico. Mineral solubility data in various inorganic salt solutions were digitized and extracted from figures generated by Harvie et al. (1984). For each of the 60 solubility curves, a case-specific chemistry model and input file were generated in OLI Studio using OLI Stream Analyzer and the MSE (H 3 O + ion) public databank provided by OLI Systems. Model simulation results were exported to Microsoft Excel to calculate summary statistics and to generate graphs comparing the OLI model predictions to the solubility data. Summary statistics include residuals (model – data) and concordance (accuracy × precision, where precision is indicated by the Pearson correlation coefficient and accuracy accounts for bias and scale differential). Private databanks were not developed, and activity coefficient model regressions were not performed to improve OLI model fits to the data. Of the 41 model verification plots, 83% have a mean of the percent residuals less than or equal to 25%. Similarly, 75% display a concordance greater than or equal to 0.75. Only seven of the 41 verification plots fail to show good agreement between the model and data. Of these seven, three are relevant to the WIPP repository because they involve the Mg-OH-Cl-SO 4 -CO 3 aqueous system. The remaining four address salt solubilities at the pH extremes (strong acid and strong base). It should be noted that in two of the three Mg-OH-Cl-SO 4 -CO 3 system cases, the regressed Harvie et al. (1984) solubility curve also deviated from the data. Lack of agreement between the OLI model-predicted solubility curves and the data is attributable to one or more of the following: specific solid species are not included in the OLI MSE databank; there is significant variation among the different solubility datasets chosen by Harvie et al. (1984); the OLI MSE model’s thermodynamic parameters were determined using different solubility datasets; and the activity coefficient parameters for certain relevant ion-ion and ion-molecule pairs have not been optimized via data regression. Two recommendations for future work are to (1) evaluate solubility data for the Mg-OH-Cl-SO 4 -CO 3 system at high ionic strength and, if necessary, develop a private OLI MSE database that includes missing species and, where necessary, regressed standard state properties and interaction parameters; (2) perform similar verification testing of the OLI model for actinide solubility data.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2020

Faster Johnson–Lindenstrauss transforms via Kronecker products

The Kronecker product is an important matrix operation with a wide range of applications in signal processing, graph theory, quantum computing and deep learning. In this work, we introduce a generalization of the fast Johnson–Lindenstrauss projection for embedding vectors with Kronecker product structure, the Kronecker fast Johnson–Lindenstrauss transform (KFJLT). The KFJLT reduces the embedding cost by an exponential factor of the standard fast Johnson–Lindenstrauss transform’s cost when applied to vectors with Kronecker structure, by avoiding explicitly forming the full Kronecker products. Here, we prove that this computational gain comes with only a small price in embedding power: consider a finite set of $p$ points in a tensor product of $d$ constituent Euclidean spaces $\bigotimes _{k=d}^{1}{\mathbb{R}}^{n_k}$, and let $N = \prod _{k=1}^{d}n_k$. With high probability, a random KFJLT matrix of dimension $m \times N$ embeds the set of points up to multiplicative distortion $(1\pm \varepsilon )$ provided $m \gtrsim \varepsilon ^{-2} \, \log ^{2d - 1} (p) \, \log N$. We conclude by describing a direct application of the KFJLT to the efficient solution of large-scale Kronecker-structured least squares problems for fitting the CP tensor decomposition.

Kronecker structure↗
DOE OSTI2017

Monte Carlo Estimates of Eigenvalue Sensitivity to System Dimensions using Kernel Density Estimators

In the last decade there has been an increasing interest in calculating sensitivity coefficients in order to quantify the impact of uncertainty in material properties and system parameters on quantities of interest. Extensive work has been done developing Monte Carlo methods to estimate eigenvalue sensitivity to material properties. While deterministic methods exist for determining the eigenvalue sensitivity to system dimensions, the use of Monte Carlo methods to efficiently calculate eigenvalue sensitivity coefficients to system dimensions is still an open topic. Current Monte Carlo methods for calculating eigenvalue sensitivity coefficients to system dimensions are inefficient. Direct Monte Carlo perturbations require two additional simulations of high statistical precision in order to determine the eigenvalue sensitivity to the location of a single boundary. If the sensitivity to the location of multiple surfaces are desired, two additional simulations must be run for each surface. Kiedrowski and Favorite extended continuous energy Monte Carlo methods for sensitivities of nuclear cross sections to calculate sensitivities to interface boundaries or system dimensions to create the only current alternative to direct Monte Carlo calculations. However this method relies on the simulation of additional random walks launched at each surface crossing event. The present method extends upon Kiedrowski's work and eliminates the inefficient portions of the calculation through the use of Kernel Density Estimators (KDEs). The new method efficiently estimates eigenvalue sensitivities to the uniform expansion or contraction of material interface boundaries or system dimensions, negligibly impacting the runtime performance of the Monte Carlo code. The new estimator is tested on bare and reflected critical configurations in spherical geometry and is compared to estimates obtained from direct Monte Carlo perturbations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Structural constraint integration in a generative model for the discovery of quantum materials

Billions of organic molecules have been computationally generated, yet functional inorganic materials remain scarce due to limited data and structural complexity. Here, in this work, we introduce Structural Constraint Integration in a GENerative model (SCIGEN), a framework that enforces geometric constraints, such as honeycomb and kagome lattices, within diffusion-based generative models to discover stable quantum materials candidates. SCIGEN enables conditional sampling from the original distribution, preserving output validity while guiding structural motifs. This approach generates ten million inorganic compounds with Archimedean and Lieb lattices, over 10% of which pass multistage stability screening. High-throughput density functional theory calculations on 26,000 candidates shows over 95% convergence and 53% structural stability. A graph neural network classifier detects magnetic ordering in 41% of relaxed structures. Furthermore, we synthesize and characterize two predicted materials, TiPd 0.22 Bi 0.88 and Ti 0.5 Pd 1.5 Sb, which display paramagnetic and diamagnetic behaviour, respectively. Our results indicate that SCIGEN provides a scalable path for generating quantum materials guided by lattice geometry.

36 MATERIALS SCIENCE↗
NASA NTRS2004

Depth-Penetrating Measurements Developed for Thermal Barrier Coatings Incorporating Thermographic Phosphors

The insulating properties of thermal barrier coatings (TBCs) provide highly beneficial thermal protection to turbine engine components by reducing the temperature sustained by those components. Therefore, measuring the temperature beneath the TBC is critical for determining whether the TBC is performing its insulating function. Currently, noncontact temperature measurements are performed by infrared pyrometry, which unfortunately measures the TBC surface temperature rather than the temperature of the underlying component. To remedy this problem, the NASA Glenn Research Center, under the Information Rich Test Instrumentation Project, developed a technique to measure the temperature beneath the TBC by incorporating a thin phosphor layer beneath the TBC. By performing fluorescence decay-time measurements on light emission from this phosphor layer, Glenn successfully measured temperatures from the phosphor layer up to 1100 C. This is the first successful demonstration of temperature measurements that penetrate beneath the TBC. Thermographic phosphors have a history of providing noncontact surface temperature measurements. Conventionally, a thermographic phosphor is applied to the material surface and temperature measurements are performed by exciting the phosphor with ultraviolet light and then measuring the temperature-dependent decay time of the phosphor emission at a longer wavelength. The innovative feature of the new approach is to take advantage of the relative transparency of the TBC (composed of yttria-stabilized zirconia) in order to excite and measure the phosphor emission beneath the TBC. The primary obstacle to achieving depth-penetrating temperature measurements is that the TBCs are completely opaque to the ultraviolet light usually employed to excite the phosphor. The strategy that Glenn pursued was to select a thermographic phosphor that could be excited and emit at wavelengths that could be transmitted through the TBC. The phosphor that was selected was yttria doped with europia (Y2O3:Eu), which has a minor excitation peak at 532 nm (green) and an emission peak at 611 nm (red)--both are wavelengths that exhibit significant transmission through the TBC. The measurements were performed on specimens consisting of a 25- m-thick phosphor layer beneath a 100- m-thick TBC. The 532-nm (green) excitation light was provided by a frequency-doubled YAG:Nd (yttrium-aluminum-garnet:neodymium) laser, and the fluorescence decay time measurements were acquired with a modified Raman microscope. The preceding graph compares the intensity of the phosphor emission of the phosphor layer above the TBC versus that of the phosphor layer beneath the TBC. Although there was considerable attenuation of the phosphor signal (a factor of 30), the phosphor emission at the reduced intensity was more than sufficient to perform fluorescence decay time measurements. The following graph shows the fluorescence lifetime temperature dependency for the Y2O3:Eu phosphor layers both above and below the TBC. These curves show an excellent match and indicate that, despite the attenuation due to the overlying TBC, the phosphor layer beneath the TBC still functions as an effective temperature indicator.

Eldridge, Jeffrey I.↗
DOE OSTI2023

Similarity Downselection: Finding the n Most Dissimilar Molecular Conformers for Reference-Free Metabolomics

Computational methods for creating in silico libraries of molecular descriptors (e.g., collision cross sections) are becoming increasingly prevalent due to the limited number of authentic reference materials available for traditional library building. These so-called “reference-free metabolomics” methods require sampling sets of molecular conformers in order to produce high accuracy property predictions. Due to the computational cost of the subsequent calculations for each conformer, there is a need to sample the most relevant subset and avoid repeating calculations on conformers that are nearly identical. The goal of this study is to introduce a heuristic method of finding the most dissimilar conformers from a larger population in order to help speed up reference-free calculation methods and maintain a high property prediction accuracy. Finding the set of the n items most dissimilar from each other out of a larger population becomes increasingly difficult and computationally expensive as either n or the population size grows large. Because there exists a pairwise relationship between each item and all other items in the population, finding the set of the n most dissimilar items is different than simply sorting an array of numbers. For instance, if you have a set of the most dissimilar n = 4 items, one or more of the items from n = 4 might not be in the set n = 5. An exact solution would have to search all possible combinations of size n in the population exhaustively. We present an open-source software called similarity downselection (SDS), written in Python and freely available on GitHub. SDS implements a heuristic algorithm for quickly finding the approximate set(s) of the n most dissimilar items. We benchmark SDS against a Monte Carlo method, which attempts to find the exact solution through repeated random sampling. We show that for SDS to find the set of n most dissimilar conformers, our method is not only orders of magnitude faster, but it is also more accurate than running Monte Carlo for 1,000,000 iterations, each searching for set sizes n = 3–7 out of a population of 50,000. We also benchmark SDS against the exact solution for example small populations, showing that SDS produces a solution close to the exact solution in these instances. Using theoretical approaches, we also demonstrate the constraints of the greedy algorithm and its efficacy as a ratio to the exact solution.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2003

New Active Optical Technique Developed for Measuring Low-Earth-Orbit Atomic Oxygen Erosion of Polymers

Polymers such as polyimide Kapton (DuPont) and Teflon FEP (DuPont, fluorinated ethylene propylene) are commonly used spacecraft materials because of desirable properties such as flexibility, low density, and in the case of FEP, a low solar absorptance and high thermal emittance. Polymers on the exterior of spacecraft in the low-Earth-orbit (LEO) environment are exposed to energetic atomic oxygen. Atomic oxygen reaction with polymers causes erosion, which is a threat to spacecraft performance and durability. It is, therefore, important to understand the atomic oxygen erosion yield E (the volume loss per incident oxygen atom) of polymers being considered in spacecraft design. The most common technique for determining E is a passive technique based on mass-loss measurements of samples exposed to LEO atomic oxygen during a space flight experiment. There are certain disadvantages to this technique. First, because it is passive, data are not obtained until after the flight is completed. Also, obtaining the preflight and postflight mass measurements is complicated by the fact that many polymers absorb water and, therefore, the mass change due to water absorption can affect the E data. This is particularly true for experiments that receive low atomic oxygen exposures or for samples that have a very low E. An active atomic oxygen erosion technique based on optical measurements has been developed that has certain advantages over the mass-loss technique. This in situ technique can simultaneously provide the erosion yield data on orbit and the atomic oxygen exposure fluence, which is needed for erosion yield determination. In the optical technique, either sunlight or artificial light can be used to measure the erosion of semitransparent or opaque polymers as a result of atomic oxygen attack. The technique is simple and adaptable to a rather wide range of polymers, providing that they have a sufficiently high optical absorption coefficient. If one covers a photodiode with a uniformly thick sheet of semitransparent polymer such as Kapton H polyimide, then as atomic oxygen erodes the polymer, the short-circuit current from the photodiode will increase in an exponential manner with fluence. This nonlinear response with fluence results in a lack of sensitivity for measuring low atomic oxygen fluences. However, if one uses a variable-thickness polymer or carbon sample, which is configured as shown in the preceding figure, then a linear response can be achieved for opaque materials using a parabolic well for a circular geometry detector or a V-shaped well for a rectangular-geometry detector. Variable-thickness samples can be fabricated using many thin polymer layers. For semitransparent polymers such as Kapton H polyimide, there is an initial short-circuit current that is greater than zero. This current has a slightly nonlinear dependence on atomic oxygen fluence in comparison to opaque materials such as black Kapton as shown in the graph. For this graph figure, the total thickness of Kapton H was assumed to be 0.03 cm. The photodiode short-circuit current shown in the graph was generated on the basis of preliminary measurements-a total reflectance rho of 0.0424 and an optical absorption coefficient a of 146.5 cm(sup -1). In addition to obtaining on-orbit data, the advantage of this active erosion and erosion yield measurement technique is its simplicity and reliance upon well-characterized fluence witness materials as well as a nearly linear photodiode short-circuit current dependence upon atomic oxygen fluence. The optical technique is useful for measuring either atomic oxygen fluence or erosion, depending on the information desired. To measure the atomic oxygen erosion yield of a test material, one would need to have two photodiode sensors, one for the test material and one that uses a known erosion yield material (such as Kapton) to measure the atomic oxygen fluence.

Banks, Bruce A.↗
DOE OSTI2019

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↗