Path Sensitive Cache Analysis Using Cache Miss Paths

Cache analysis plays a very important role in obtaining precise Worst Case Execution Time (WCET) estimates of programs for real-time systems. While Abstract Interpretation based approaches are almost universally used for cache analysis, they fail to take advantage of its unique requirement: it is not necessary to find the guaranteed cache behavior that holds across all executions of a program. We only need the cache behavior along one particular program path, which is the path with the maximum execution time. In this work, we introduce the concept of cache miss paths, which allows us to use the worst-case path information to improve the precision of AI-based cache analysis. We use Abstract Interpretation to determine the cache miss paths, and then integrate them in the IPET formulation. An added advantage is that this further allows us to use infeasible path information for cache analysis. Experimentally, our approach gives more precise WCETs as compared to AI-based cache analysis, and we also provide techniques to trade-off analysis time with precision to provide scalability.

Phase space path integral approach to harmonic oscillator with a time-dependent force constant

The quantum statistical mechanical propagator for a harmonic oscillator with a time-dependent force constant, m omega(2)(t), has been investigated in the past and was found to have only a formal solution in terms of the solutions of certain ordinary differential equations. Such path integrals are frequently encountered in semiclassical path integral evaluations and having exact analytical expressions for such path integrals is of great interest. In a previous work, we had obtained the exact propagator for motion in an arbitrary time-dependent harmonic potential in the overdamped limit of friction using phase space path integrals in the context of Levy flights - a result that can be easily extended to Brownian motion. In this paper, we make a connection between the overdamped Brownian motion and the imaginary time propagator of quantum mechanics and thereby get yet another way to evaluate the latter exactly. We find that explicit analytic solution for the quantum statistical mechanical propagator can be written when the time-dependent force constant has the form omega(2)(t) = lambda(2)(t) - d lambda(t)/dt where lambda(t) is any arbitrary function of t and use it to evaluate path integrals which have not been evaluated previously. We also employ this method to arrive at a formal solution of the propagator for both Levy flights and Brownian subjected to a time-dependent harmonic potential in the underdamped limit of friction. (C) 2015 Elsevier B.V. All rights reserved.

Taming Control Divergence in GPUs through Control Flow Linearization

Branch divergence is a very commonly occurring performance problem in GPGPU in which the execution of diverging branches is serialized to execute only one control flow path at a time. Existing hardware mechanism to reconverge threads using a stack causes duplicate execution of code for unstructured control flow graphs. Also the stack mechanism cannot effectively utilize the available parallelism among diverging branches. Further, the amount of nested divergence allowed is also limited by depth of the branch divergence stack. In this paper we propose a simple and elegant transformation to handle all of the above mentioned problems. The transformation converts an unstructured CFG to a structured CFG without duplicating user code. It incurs only a linear increase in the number of basic blocks and also the number of instructions. Our solution linearizes the CFG using a predicate variable. This mechanism reconverges the divergent threads as early as possible. It also reduces the depth of the reconvergence stack. The available parallelism in nested branches can be effectively extracted by scheduling the basic blocks to reduce the effect of stalls due to memory accesses. It can also increase execution efficiency of nested loops with different trip counts for different threads. We implemented the proposed transformation at PTX level using the Ocelot compiler infrastructure. We evaluated the technique using various benchmarks to show that it can be effective in handling the performance problem due to divergence in unstructured CFGs.

Ag+-induced reverse vesicle to helical fiber transformation in a self-assembly by adjusting the keto-enol equilibrium of a chiral salicylideneaniline

A new chiral amphiphilic salicylideneaniline bearing a terminal pyridine was synthesized. It formed reverse vesicles in toluene. The addition of Ag+, however, reversibly transforms these reverse vesicles into left-handed nanohelices accompanied by spontaneous gel formation at room temperature.

Dependence of shear yield strain and shear transformation zone on the glass transition temperature in thin film metallic glasses

The dependence of shear yield strain, the activation energy and volume of shear transformation zone on the glass transition temperature was investigated through the analysis of statistical distributions of the first pop-in events during spherical indentation of four different thin film metallic glasses. Only the Cu-Zr metallic glass exhibits a bimodal distribution of the first pop-in loads, whereas W-Ru-B, Zr-Cu-Ni-Al and La-Co-Al metallic glasses show an unimodal distribution. Results show that shear yield strain and activation energy of shear transformation zone decrease whereas the volume of shear transformation zone increases with increasing homologous temperature, indicating that it is the activation energy rather than the volume of shear transformation zone that controls shear yield strain. (C) 2015 Elsevier B.V. All rights reserved.

Partial rotation of the isopropyl group in the solid state: single-crystal-to-single-crystal phase transformation in a carvacrol derivative

A hitherto unseen rotation of the isopropyl group in the solid state, predicted to be forbidden based on theoretical investigations, is reported. This C-C rotation observed during the temperature dependent single-crystal-to-single-crystal transformation is attributed to the concomitant changes in molecular structure and intermolecular packing.

Ray tracing based path-length calculations for polarized light tomographic imaging

A ray tracing based path length calculation is investigated for polarized light transport in a pixel space. Tomographic imaging using polarized light transport is promising for applications in optical projection tomography of small animal imaging and turbid media with low scattering. Polarized light transport through a medium can have complex effects due to interactions such as optical rotation of linearly polarized light, birefringence, diattenuation and interior refraction. Here we investigate the effects of refraction of polarized light in a non-scattering medium. This step is used to obtain the initial absorption estimate. This estimate can be used as prior in Monte Carlo (MC) program that simulates the transport of polarized light through a scattering medium to assist in faster convergence of the final estimate. The reflectance for p-polarized (parallel) and s-polarized (perpendicular) are different and hence there is a difference in the intensities that reach the detector end. The algorithm computes the length of the ray in each pixel along the refracted path and this is used to build the weight matrix. This weight matrix with corrected ray path length and the resultant intensity reaching the detector for each ray is used in the algebraic reconstruction (ART) method. The proposed method is tested with numerical phantoms for various noise levels. The refraction errors due to regions of different refractive index are discussed, the difference in intensities with polarization is considered. The improvements in reconstruction using the correction so applied is presented. This is achieved by tracking the path of the ray as well as the intensity of the ray as it traverses through the medium.

Relaxor ferroelectricity and electric-field-driven structural transformation in the giant lead-free piezoelectric (Ba,Ca)(Ti, Zr)O-3

There is great interest in lead-free (Ba0.85Ca0.15)(Ti0.90Zr0.10)O-3 (15/10BCTZ) because of its exceptionally large piezoelectric response Liu and Ren, Phys. Rev. Lett. 103, 257602 (2009)]. In this paper, we have analyzed the nature of: (i) crystallographic phase coexistence at room temperature, (ii) temperature-and field-induced phase transformation to throw light on the atomistic mechanisms associated with the large piezoelectric response of this system. A detailed temperature-dependent dielectric and lattice thermal expansion study proved that the system exhibits a weak dielectric relaxation, characteristic of a relaxor ferroelectric material on the verge of exhibiting a normal ferroelectric-paraelectric transformation. Careful structural analysis revealed that a ferroelectric state at room temperature is composed of three phase coexistences, tetragonal (P4mm)+ orthorhombic (Amm2) + rhombohedral (R3m). We also demonstrate that the giant piezoresponse is associated with a significant fraction of the tetragonal phase transforming to rhombohedral. It is argued that the polar nanoregions associated with relaxor ferroelectricity amplify the piezoresponse by providing an additional degree of intrinsic structural inhomogeneity to the system.

Monoclinic to cubic phase transformation in combustion synthesized gadolinium oxide

Gadolinium oxide, cerium oxide, and 10 mol% gadolinia doped ceria ceramic powders have been synthesized using combustion technique. Though the cubic gadolinia phase is stable at room temperature, single phase monoclinic gadolinia was obtained as a result of combustion synthesis using fuel lean and stoichiometric precursor compositions. This powder was subjected to calcination treatment and ceria doping to study the stability of phases and the rate of phase transformation from monoclinic to cubic gadolinia. It was found that monoclinic gadolinia transforms to cubic gadolinia upon calcination at temperatures less than 1200 degrees C. It was also found that rate of phase transformation is more for powder produced using fuel lean compositions; and the rate is enhanced upon ceria doping. (C) 2015 Elsevier Ltd. All rights reserved.

A coupled discriminative dictionary and transformation learning approach with applications to cross domain matching

Cross domain and cross-modal matching has many applications in the field of computer vision and pattern recognition. A few examples are heterogeneous face recognition, cross view action recognition, etc. This is a very challenging task since the data in two domains can differ significantly. In this work, we propose a coupled dictionary and transformation learning approach that models the relationship between the data in both domains. The approach learns a pair of transformation matrices that map the data in the two domains in such a manner that they share common sparse representations with respect to their own dictionaries in the transformed space. The dictionaries for the two domains are learnt in a coupled manner with an additional discriminative term to ensure improved recognition performance. The dictionaries and the transformation matrices are jointly updated in an iterative manner. The applicability of the proposed approach is illustrated by evaluating its performance on different challenging tasks: face recognition across pose, illumination and resolution, heterogeneous face recognition and cross view action recognition. Extensive experiments on five datasets namely, CMU-PIE, Multi-PIE, ChokePoint, HFB and IXMAS datasets and comparisons with several state-of-the-art approaches show the effectiveness of the proposed approach. (C) 2015 Elsevier B.V. All rights reserved.

A coupled discriminative dictionary and transformation learning approach with applications to cross domain matching

Cross domain and cross-modal matching has many applications in the field of computer vision and pattern recognition. A few examples are heterogeneous face recognition, cross view action recognition, etc. This is a very challenging task since the data in two domains can differ significantly. In this work, we propose a coupled dictionary and transformation learning approach that models the relationship between the data in both domains. The approach learns a pair of transformation matrices that map the data in the two domains in such a manner that they share common sparse representations with respect to their own dictionaries in the transformed space. The dictionaries for the two domains are learnt in a coupled manner with an additional discriminative term to ensure improved recognition performance. The dictionaries and the transformation matrices are jointly updated in an iterative manner. The applicability of the proposed approach is illustrated by evaluating its performance on different challenging tasks: face recognition across pose, illumination and resolution, heterogeneous face recognition and cross view action recognition. Extensive experiments on five datasets namely, CMU-PIE, Multi-PIE, ChokePoint, HFB and IXMAS datasets and comparisons with several state-of-the-art approaches show the effectiveness of the proposed approach. (C) 2015 Elsevier B.V. All rights reserved.

Smooth Path Planning for Passages with Heading and Curvature Discontinuities

The problem of continuous curvature path planning for passages is considered. This problem arises when an autonomous vehicle traverses between prescribed boundaries such as corridors, tunnels, channels, etc. Passage boundaries with curvature and heading discontinuities pose challenges for generating smooth paths passing through them. Continuous curvature half-S shaped paths derived from the Four Parameter Logistic Curve family are proposed as a prospective path planning solution. Analytic conditions are derived for generating continuous curvature paths confined within the passage boundaries. Zero end curvature highlights the scalability of the proposed solution and its compatibility with other path planners in terms of larger path planning domains. Various scenarios with curvature and heading discontinuities are considered presenting viability of the proposed solution.

Long-period modulated structure and electric-field-induced structural transformation in Na0.5Bi0.5TiO3-based lead-free piezoelectrics

Na0.5Bi0.5TiO3- based lead-free piezoelectrics exhibiting giant piezostrain are technologically interesting materials for actuator applications. The lack of clarity with regard to the structure of the nonpolar phase of this system has hindered the understanding of the structural mechanism associated with the giant piezostrain and other related phenomena. In this paper, we have investigated the structure and field-induced phase transformation behavior of a model system (0.94 - x) Na0.5Bi0.5TiO3-0.06BaTiO(3)-xK(0.5)Na(0.5)NbO(3) (0.0 <= x <= 0.025). A detailed structural analysis using neutron powder diffraction revealed that the nonpolar phase is neither cubic nor a mixture of rhombohedral (R3c) and tetragonal (P4bm) phases as commonly reported in literature but exhibits a long-period modulated structure, which is most probably of the type root 2 x root 2 x n with n = 16. Our results suggest that the giant piezoelectric strain is associated with a field-induced phase transformation of the long-period modulated structure to rhombohedral R3c structure above a critical field. We also demonstrate that the giant piezostrain is lost if the system retains a fraction of the field-induced R3c phase. A possible correlation among depolarization temperature, giant piezostrain, and its electrical fatigue behavior has also been indicated.