Focused Prefetching: Performance Oriented Prefetching Based On Commit Stalls

Loads that miss in L1 or L2 caches and waiting for their data at the head of the ROB cause significant slow down in the form of commit stalls. We identify that most of these commit stalls are caused by a small set of loads, referred to as LIMCOS (Loads Incurring Majority of COmmit Stalls). We propose simple history-based classifiers that track commit stalls suffered by loads to help us identify this small set of loads. We study an application of these classifiers to prefetching. The classifiers are used to train the prefetcher to focus on the misses suffered by LIMCOS. This, referred to as focused prefetching, results in a 9.8% gain in IPC over naive GHB based delta correlation prefetcher along with a 20.3% reduction in memory traffic for a set of 17 memory-intensive SPEC2000 benchmarks. Another important impact of focused prefetching is a 61% improvement in the accuracy of prefetches. We demonstrate that the proposed classification criterion performs better than other existing criteria like criticality and delinquent loads. Also we show that the criterion of focusing on commit stalls is robust enough across cache levels and can be applied to any prefetcher without any modifications to the prefetcher.

On the determination of energy eigenvalues and coupling strengths using duality

The Shifman-Vainshtein-Zakharov method of determining the eigenvalues and coupling strengths, from the operator product expansion, for the current correlation functions is studied in the nonrelativistic context, using the semiclassical expansion. The relationship between the low-lying eigenvalues, and the leading corrections to the imaginary-time Green function is elucidated by comparing systems which have almost identical spectra. In the case of an anharmonic oscillator it is found that with the procedure stated in the paper, that inclusion of more terms to the asymptotic expansion does not show any simple trend towards convergence to the exact values. Generalization to higher partial waves is given. In particular for the P-level of the oscillator, the procedure gives poorer results than for the S-level, although the ratio of the two comes out much better.

A Simple Method of Obtaining the Band Strengths in the Electronic Spectra of Diatomic Molecules

Relative band strengths of diatomic molecules for which the product of Franck-Condon factor and r-centroid is approximately equal to 1 for (0,0) band can be determined by a simple method which will be in good agreement with the smoothed array of experimental values.

Calculation of Coulomb interaction strengths for 3d transition metals and actinides

Coulomb interaction strengths (Udd and Uff) have been calculated from Hartree-Fock-Slater atomic calculations for 3d transition and 5f actinide elements, respectively. By decomposing the different contributions to the response (screening) to the 3d charge fluctuation, we show that a substantial reduction in Udd arises due to the relaxation of the 3d charge distribution itself. This, combined with the screening due to the response of the 4s charge density, is shown to provide a very compact screening charge comparable to the metallic case, explaining the success of the atomic calculations for estimating U even in the metals. A pronounced dependence of Udd (or Uff) on the number of electrons nd (nf) or the electronic configuration is also shown here.

Characterization of compact ICP ion source for focused ion beam applications

A compact, high brightness 13.56 MHz inductively coupled plasma ion source without any axial or radial multicusp magnetic fields is designed for the production of a focused ion beam. Argon ion current of density more than 30 mA/cm(2) at 4 kV potential is extracted from this ion source and is characterized by measuring the ion energy spread and brightness. Ion energy spread is measured by a variable-focusing retarding field energy analyzer that minimizes the errors due t divergence of ion beam inside the analyzer. Brightness of the ion beam is determined from the emittance measured by a fully automated and locally developed electrostatic sweep scanner. By optimizing various ion source parameters such as RF power, gas pressure and Faraday shield, ion beams with energy spread of less than 5 eV and brightness of 7100 Am(-2)sr(-1)eV(-1) have been produced. Here, we briefly report the details of the ion source, measurement and optimization of energy spread and brightness of the ion beam. (C) 2010 Elsevier B.V. All rights reserved.

Optimization of ion energy spread in inductively coupled plasma source designed for focused ion beam applications

The ion energy distribution of inductively coupled plasma ion source for focused ion beam application is measured using a four grid retarding field energy analyzer. Without using any Faraday shield, ion energy spread is found to be 50 eV or more. Moreover, the ion energy distribution is found to have double peaks showing that the power coupling to the plasma is not purely inductive, but a strong parasitic capacitive coupling is also present. By optimizing the various source parameters and Faraday shield, ion energy distribution having a single peak, well separated from zero energy and with ion energy spread of 4 eV is achieved. A novel plasma chamber, with proper Faraday shield is designed to ignite the plasma at low RF powers which otherwise would require 300-400 W of RF power. Optimization of various parameters of the ion source to achieve ions with very low energy spread and the experimental results are presented in this article. (C) 2010 Elsevier Ltd. All rights reserved.

Studies on the variations in compression strengths of graphite powders bearing glass-epoxy composites exposed to humid conditions

The moisture absorption and changes in compression strengths in glass-epoxy (G-E composites without and with discrete quantities of graphite powders introduced into the resin mix prior to its spreading on specific glass fabric (layers) during the lay-up (stacking) sequence forms the subject matter of this report. The results point to higher moisture absorption for graphite bearing specimens. The strengths of graphite-free coupons show a continuous decrease, while the filler bearing ones show an initial rise followed by a drop for larger exposure times. Scanning Fractographic features were examined for an understanding of the process. The observations were explained invoking the effect of matrix plasticizing and the role of interfacial regions.

Approaching theoretical strengths by synergistic internal and external size refinement

Refinement of the internal grain size leads to strengthening by retarding dislocation motion. There have also been recent reports that a reduction in external diameter enhances the strength of single crystal pillars. Here we show, in a hitherto unexplored domain, a synergistic increase in strength by a combined reduction in internal (0.5 mu m) and external (20-50 mu m) dimensions, with strengths at failure approaching the theoretical value. (c) 2012 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Improving query focused summarization using look-ahead strategy

Query focused summarization is the task of producing a compressed text of original set of documents based on a query. Documents can be viewed as graph with sentences as nodes and edges can be added based on sentence similarity. Graph based ranking algorithms which use 'Biased random surfer model' like topic-sensitive LexRank have been successfully applied to query focused summarization. In these algorithms, random walk will be biased towards the sentences which contain query relevant words. Specifically, it is assumed that random surfer knows the query relevance score of the sentence to where he jumps. However, neighbourhood information of the sentence to where he jumps is completely ignored. In this paper, we propose look-ahead version of topic-sensitive LexRank. We assume that random surfer not only knows the query relevance of the sentence to where he jumps but he can also look N-step ahead from that sentence to find query relevance scores of future set of sentences. Using this look ahead information, we figure out the sentences which are indirectly related to the query by looking at number of hops to reach a sentence which has query relevant words. Then we make the random walk biased towards even to the indirect query relevant sentences along with the sentences which have query relevant words. Experimental results show 20.2% increase in ROUGE-2 score compared to topic-sensitive LexRank on DUC 2007 data set. Further, our system outperforms best systems in DUC 2006 and results are comparable to state of the art systems.

The effect of gamma-gamma ` interface on the tensile and shear strengths of nickel-based superalloys: A first-principles study

First-principles density functional theory has been used to evaluate the shear and cleavage strength in terms of Griffith work and generalized stacking fault energy (GSF) of (001) plane for gamma, gamma' and gamma-gamma' system as a function of distance from the gamma/gamma' interface. Calculation of Griffith work suggests higher cleavage energy for bulk gamma as compared to gamma' while the GSF calculation suggests higher shear strength for bulk gamma' as compared to gamma. It has been found that the shear strength of the cubic plane of the gamma/gamma' interface is marginally lower than those of bulk gamma and gamma' phases. (C) 2014 Elsevier B.V. All rights reserved.

Prior image based temporally constrained reconstruction algorithm for magnetic resonance guided high intensity focused ultrasound

Purpose: A prior image based temporally constrained reconstruction ( PITCR) algorithm was developed for obtaining accurate temperature maps having better volume coverage, and spatial, and temporal resolution than other algorithms for highly undersampled data in magnetic resonance (MR) thermometry. Methods: The proposed PITCR approach is an algorithm that gives weight to the prior image and performs accurate reconstruction in a dynamic imaging environment. The PITCR method is compared with the temporally constrained reconstruction (TCR) algorithm using pork muscle data. Results: The PITCR method provides superior performance compared to the TCR approach with highly undersampled data. The proposed approach is computationally expensive compared to the TCR approach, but this could be overcome by the advantage of reconstructing with fewer measurements. In the case of reconstruction of temperature maps from 16% of fully sampled data, the PITCR approach was 1.57x slower compared to the TCR approach, while the root mean square error using PITCR is 0.784 compared to 2.815 with the TCR scheme. Conclusions: The PITCR approach is able to perform more accurate reconstructions of temperature maps compared to the TCR approach with highly undersampled data in MR guided high intensity focused ultrasound. (C) 2015 American Association of Physicists in Medicine.