PLUTO plus : Near-Complete Modeling of Affine Transformations for Parallelism and Locality

Affine transformations have proven to be very powerful for loop restructuring due to their ability to model a very wide range of transformations. A single multi-dimensional affine function can represent a long and complex sequence of simpler transformations. Existing affine transformation frameworks like the Pluto algorithm, that include a cost function for modern multicore architectures where coarse-grained parallelism and locality are crucial, consider only a sub-space of transformations to avoid a combinatorial explosion in finding the transformations. The ensuing practical tradeoffs lead to the exclusion of certain useful transformations, in particular, transformation compositions involving loop reversals and loop skewing by negative factors. In this paper, we propose an approach to address this limitation by modeling a much larger space of affine transformations in conjunction with the Pluto algorithm's cost function. We perform an experimental evaluation of both, the effect on compilation time, and performance of generated codes. The evaluation shows that our new framework, Pluto+, provides no degradation in performance in any of the Polybench benchmarks. For Lattice Boltzmann Method (LBM) codes with periodic boundary conditions, it provides a mean speedup of 1.33x over Pluto. We also show that Pluto+ does not increase compile times significantly. Experimental results on Polybench show that Pluto+ increases overall polyhedral source-to-source optimization time only by 15%. In cases where it improves execution time significantly, it increased polyhedral optimization time only by 2.04x.

Models for the estimation of life of stator insulation in large rotating machines

Epoxy resin bonded mica splitting is the insulation of choice for machine stators. However, this system is seen to be relatively weak under time varying mechanical stress, in particular the vibration causing delamination of mica and deboning of mica from the resin matrix. The situation is accentuated under the combined action of electrical, thermal and mechanical stress. Physical and probabilistic models for failure of such systems have been proposed by one of the authors of this paper earlier. This paper presents a pragmatic accelerated failure data acquisition and analytical paradigm under multi factor coupled stress, Electrical, Thermal. The parameters of the phenomenological model so developed are estimated based on sound statistical treatment of failure data.

A Gaussian network model study suggests that structural fluctuations are higher for inactive states than active states of protein kinases

We performed Gaussian network model based normal mode analysis of 3-dimensional structures of multiple active and inactive forms of protein kinases. In 14 different kinases, a more number of residues (1095) show higher structural fluctuations in inactive states than those in active states (525), suggesting that, in general, mobility of inactive states is higher than active states. This statistically significant difference is consistent with higher crystallographic B-factors and conformational energies for inactive than active states, suggesting lower stability of inactive forms. Only a small number of inactive conformations with the DFG motif in the ``in'' state were found to have fluctuation magnitudes comparable to the active conformation. Therefore our study reports for the first time, intrinsic higher structural fluctuation for almost all inactive conformations compared to the active forms. Regions with higher fluctuations in the inactive states are often localized to the aC-helix, aG-helix and activation loop which are involved in the regulation and/or in structural transitions between active and inactive states. Further analysis of 476 kinase structures involved in interactions with another domain/protein showed that many of the regions with higher inactive-state fluctuation correspond to contact interfaces. We also performed extensive GNM analysis of (i) insulin receptor kinase bound to another protein and (ii) holo and apo forms of active and inactive conformations followed by multi-factor analysis of variance. We conclude that binding of small molecules or other domains/proteins reduce the extent of fluctuation irrespective of active or inactive forms. Finally, we show that the perceived fluctuations serve as a useful input to predict the functional state of a kinase.

Over-expression and purification strategies for recombinant multi-protein oligomers: A case study of Mycobacterium tuberculosis sigma/anti-sigma factor protein complexes

The function of a protein in a cell often involves coordinated interactions with one or several regulatory partners. It is thus imperative to characterize a protein both in isolation as well as in the context of its complex with an interacting partner. High resolution structural information determined by X-ray crystallography and Nuclear Magnetic Resonance offer the best route to characterize protein complexes. These techniques, however, require highly purified and homogenous protein samples at high concentration. This requirement often presents a major hurdle for structural studies. Here we present a strategy based on co-expression and co-purification to obtain recombinant multi-protein complexes in the quantity and concentration range that can enable hitherto intractable structural projects. The feasibility of this strategy was examined using the sigma factor/anti-sigma factor protein complexes from Mycobacterium tuberculosis. The approach was successful across a wide range of sigma factors and their cognate interacting partners. It thus appears likely that the analysis of these complexes based on variations in expression constructs and procedures for the purification and characterization of these recombinant protein samples would be widely applicable for other multi-protein systems. (C) 2010 Elsevier Inc. All rights reserved.

MPI-based parallel synchronous vector evaluated particle swarm optimization for multi-objective design optimization of composite structures

This paper presents a decentralized/peer-to-peer architecture-based parallel version of the vector evaluated particle swarm optimization (VEPSO) algorithm for multi-objective design optimization of laminated composite plates using message passing interface (MPI). The design optimization of laminated composite plates being a combinatorially explosive constrained non-linear optimization problem (CNOP), with many design variables and a vast solution space, warrants the use of non-parametric and heuristic optimization algorithms like PSO. Optimization requires minimizing both the weight and cost of these composite plates, simultaneously, which renders the problem multi-objective. Hence VEPSO, a multi-objective variant of the PSO algorithm, is used. Despite the use of such a heuristic, the application problem, being computationally intensive, suffers from long execution times due to sequential computation. Hence, a parallel version of the PSO algorithm for the problem has been developed to run on several nodes of an IBM P720 cluster. The proposed parallel algorithm, using MPI's collective communication directives, establishes a peer-to-peer relationship between the constituent parallel processes, deviating from the more common master-slave approach, in achieving reduction of computation time by factor of up to 10. Finally we show the effectiveness of the proposed parallel algorithm by comparing it with a serial implementation of VEPSO and a parallel implementation of the vector evaluated genetic algorithm (VEGA) for the same design problem. (c) 2012 Elsevier Ltd. All rights reserved.

pH sensing by single and multi-layer hydrogel coated Fiber Bragg Grating

In this paper we show a novel chemo-mechanical-optical sensing mechanism in single and multi-layer hydrogel coated Fiber Bragg Grating (FBG) and demonstrate specific application in pH activated processes. The sensing device is based on the ionizable monomers inside the hydrogel which reversibly dissociates as a function of the pH and consequently resulting in osmotic pressure difference between the gel and the solution. This pressure gradient causes the hydrogel to deform which in turn induces secondary strain on the FBG sensor resulting in shift in the Bragg wavelength. We also report on the sensitivity factor of single and multilayer hydrogel coated FBG at various different pH.

Pressure sensitive flow and constraint factor in amorphous materials below glass transition

The constraint factor, C (given by the hardness-yield strength ratio H/Y in the fully lastic regime of indentation), in metallic glasses, is greater than three, a reflection of the sensitivity of their plastic flow to pressure. Furthermore, C increases with increasing temperature. In this work, we examine if this is true in amorphous polymers as well, through experiments on amorphous poly(methyl methacrylate) (PMMA). Uniaxial compression as well as spherical indentation tests were conducted in the 248-348 K range to construct H/Y versus indentation strain plots at each temperature and obtain the C-values. Results show that C increases with temperature in PMMA as well. Good correlation between the loss factors, measured using a dynamic mechanical analyzer, and C, suggest that the enhanced sensitivity to pressure is possibly due to beta-relaxation. We offer possible mechanistic reasons for the observed trends in amorphous materials in terms of relaxation processes.

Statistical significance of sequential firing patterns in multi-neuronal spike trains

Sequential firings with fixed time delays are frequently observed in simultaneous recordings from multiple neurons. Such temporal patterns are potentially indicative of underlying microcircuits and it is important to know when a repeatedly occurring pattern is statistically significant. These sequences are typically identified through correlation counts. In this paper we present a method for assessing the significance of such correlations. We specify the null hypothesis in terms of a bound on the conditional probabilities that characterize the influence of one neuron on another. This method of testing significance is more general than the currently available methods since under our null hypothesis we do not assume that the spiking processes of different neurons are independent. The structure of our null hypothesis also allows us to rank order the detected patterns. We demonstrate our method on simulated spike trains.

Metamodeling approach to predict friction factor of alluvial channel

The flow resistance of an alluvial channel flow is not only affected by the Reynolds number and the roughness conditions but also the Froude number. Froude number is the most basic parameter in the case of the alluvial channel, thus effect of Froude number on resistance to flow should be considered in the formulation of the friction factor, which is not in the case of present available resistance equations. At present, no generally acceptable quantitative description of the effects of the Froude number on hydraulic resistance has been developed. Metamodeling technique, which is particularly useful in modeling a complex processes or where knowledge of the physics is limited, is presented as a tool complimentary to modeling friction factor in alluvial channels. Present work uses, a radial basis metamodel, which is a type of neural network modeling, to find the effect of Froude number on the flow resistance. Based on the experimental data taken from different sources, it has been found that the predicting capability of the present model is on acceptable level. Present work also tries in formulating an empirical equation for resistance in alluvial channel comprising all the three majorm, parameters, namely, roughness parameter, Froude number and Reynolds number. (C) 2009 Elsevier B.V. All rights reserved.

DMT of Multi-hop Cooperative Networks-Part II: Layered and Multi-Antenna Networks

We consider single-source, single-sink (ss-ss) multi-hop relay networks, with slow-fading Rayleigh links. This two part paper aims at giving explicit protocols and codes to achieve the optimal diversity-multiplexing tradeoff (DMT) of two classes of multi-hop networks: K-parallel-path (KPP) networks and Layered networks. While single-antenna KPP networks were the focus of the first part, we consider layered and multi-antenna networks in this second part. We prove that a linear DMT between the maximum diversity d(max). and the maximum multiplexing gain of 1 is achievable for single-antenna fully-connected layered networks under the half-duplex constraint. This is shown to be equal to the optimal DMT if the number of relaying layers is less than 4. For the multiple-antenna case, we provide an achievable DMT, which is significantly better than known lower bounds for half duplex networks. Along the way, we compute the DMT of parallel MIMO channels in terms of the DMT of the component channel. For arbitrary ss-ss single-antenna directed acyclic networks with full-duplex relays, we prove that a linear tradeoff between maximum diversity and maximum multiplexing gain is achievable using an amplify-and-forward (AF) protocol. Explicit short-block-length codes are provided for all the proposed protocols. Two key implications of the results in the two-part paper are that the half-duplex constraint does not necessarily entail rate loss by a factor of two as previously believed and that simple AN protocols are often sufficient to attain the best possible DMT.

DMT of Multi-hop Cooperative Networks-Part I: K-Parallel-Path Networks

We consider single-source, single-sink multi-hop relay networks, with slow-fading Rayleigh fading links and single-antenna relay nodes operating under the half-duplex constraint. While two hop relay networks have been studied in great detail in terms of the diversity-multiplexing tradeoff (DMT), few results are available for more general networks. In this two-part paper, we identify two families of networks that are multi-hop generalizations of the two hop network: K-Parallel-Path (KPP) networks and Layered networks. In the first part, we initially consider KPP networks, which can be viewed as the union of K node-disjoint parallel paths, each of length > 1. The results are then generalized to KPP(I) networks, which permit interference between paths and to KPP(D) networks, which possess a direct link from source to sink. We characterize the optimal DMT of KPP(D) networks with K >= 4, and KPP(I) networks with K >= 3. Along the way, we derive lower bounds for the DMT of triangular channel matrices, which are useful in DMT computation of various protocols. As a special case, the DMT of two-hop relay network without direct link is obtained. Two key implications of the results in the two-part paper are that the half-duplex constraint does not necessarily entail rate loss by a factor of two, as previously believed and that, simple AF protocols are often sufficient to attain the best possible DMT.

Multi-criteria seismic hazard evaluation for Bangalore city, India

Different seismic hazard components pertaining to Bangalore city,namely soil overburden thickness, effective shear-wave velocity, factor of safety against liquefaction potential, peak ground acceleration at the seismic bedrock, site response in terms of amplification factor, and the predominant frequency, has been individually evaluated. The overburden thickness distribution, predominantly in the range of 5-10 m in the city, has been estimated through a sub-surface model from geotechnical bore-log data. The effective shear-wave velocity distribution, established through Multi-channel Analysis of Surface Wave (MASW) survey and subsequent data interpretation through dispersion analysis, exhibits site class D (180-360 m/s), site class C (360-760 m/s), and site class B (760-1500 m/s) in compliance to the National Earthquake Hazard Reduction Program (NEHRP) nomenclature. The peak ground acceleration has been estimated through deterministic approach, based on the maximum credible earthquake of M-W = 5.1 assumed to be nucleating from the closest active seismic source (Mandya-Channapatna-Bangalore Lineament). The 1-D site response factor, computed at each borehole through geotechnical analysis across the study region, is seen to be ranging from around amplification of one to as high as four times. Correspondingly, the predominant frequency estimated from the Fourier spectrum is found to be predominantly in range of 3.5-5.0 Hz. The soil liquefaction hazard assessment has been estimated in terms of factor of safety against liquefaction potential using standard penetration test data and the underlying soil properties that indicates 90% of the study region to be non-liquefiable. The spatial distributions of the different hazard entities are placed on a GIS platform and subsequently, integrated through analytical hierarchal process. The accomplished deterministic hazard map shows high hazard coverage in the western areas. The microzonation, thus, achieved is envisaged as a first-cut assessment of the site specific hazard in laying out a framework for higher order seismic microzonation as well as a useful decision support tool in overall land-use planning, and hazard management. (C) 2010 Elsevier Ltd. All rights reserved.

Stability of synchronization in a multi-cellular system

Networks of biochemical reactions regulated by positive-and negative-feedback processes underlie functional dynamics in single cells. Synchronization of dynamics in the constituent cells is a hallmark of collective behavior in multi-cellular biological systems. Stability of the synchronized state is required for robust functioning of the multi-cell system in the face of noise and perturbation. Yet, the ability to respond to signals and change functional dynamics are also important features during development, disease, and evolution in living systems. In this paper, using a coupled multi-cell system model, we investigate the role of system size, coupling strength and its topology on the synchronization of the collective dynamics and its stability. Even though different coupling topologies lead to synchronization of collective dynamics, diffusive coupling through the end product of the pathway does not confer stability to the synchronized state. The results are discussed with a view to their prevalence in biological systems. Copyright (C) EPLA, 2010

An actor-critic algorithm with function approximation for discounted cost constrained Markov decision processes

We develop in this article the first actor-critic reinforcement learning algorithm with function approximation for a problem of control under multiple inequality constraints. We consider the infinite horizon discounted cost framework in which both the objective and the constraint functions are suitable expected policy-dependent discounted sums of certain sample path functions. We apply the Lagrange multiplier method to handle the inequality constraints. Our algorithm makes use of multi-timescale stochastic approximation and incorporates a temporal difference (TD) critic and an actor that makes a gradient search in the space of policy parameters using efficient simultaneous perturbation stochastic approximation (SPSA) gradient estimates. We prove the asymptotic almost sure convergence of our algorithm to a locally optimal policy. (C) 2010 Elsevier B.V. All rights reserved.

Multi frequency interrogation of polypyrrole based gas sensors for organic vapors

Polypyrrole exhibits reversible changes in their direct current resistance on exposure to organic volatiles. However, one needs to employ an array of such sensors to discriminate organic volatiles present in a mixture. Hence, polypyrrole based gas sensor is designed for the detection and discrimination of different organic volatiles. Multi frequency impedance measurement technique is used to detect the organic vapors, such as acetone, ethanol and Isopropyl alcohol, in the gas phase, over a frequency range 10 Hz to 2 MHz. The sensor response is monitored by measuring the changes in its capacitance, resistance and the dissipation factor upon exposure to organic volatiles. It is observed that the capacitive property of the sensor is more sensitive to these volatiles than its resistive property. Each volatile responds to the sensor in terms of dissipation factor at specific frequency and found that the peak magnitude has a linear relationship with their concentrations.