Predicting x86 Program Runtime for Intel Processor
|
|
- Cecil Lenard Woods
- 5 years ago
- Views:
Transcription
1 Predicting x86 Progra Runtie for Intel Processor Behra Mistree, Haidreza Haki Javadi, Oid Mashayekhi Stanford University 1 Introduction Progras can be equivalent: given the sae set of inputs, they always produce the sae set of outputs. Superoptiizing copilers leverage this notion of equivalence, substituting one progra (or section of progra) with an equivalent progra that runs faster. Recent work by Schkufza attepts to build a superoptiizing copiler that intelligently searches a progra space coposed of all equivalent sets of instructions [3]. This work relies on being able to predict a progra s runtie given its progra text. This report explores several runtie prediction odels. Specifically, given a set of loop-free x86 input instructions, our project predicts how long a processor takes to execute those instructions. Section 2 discusses predicting runties for fixed length progras and shows that kernelized regression odels can predict progra runtie with 12% average error copared to a gold-standard tool [1]. Following these results, Section 3 proposes and evaluates a generative odel for predicting runties for progras with arbitrary nubers of instructions. Finally, Section 4 provides a perforance analysis of the proposed predicter. 2 Fixed length progras This section builds odels to predict runties for progras coposed of a fixed nubers of instructions. Section 2.1 shows the perforance of a siple linear regression and 2.2 expand this odel using Gaussian and polynoial kernels. 2.1 Linear regression Figure 1 shows the results of applying linear regression on randoly generated progras of length 1, 2, 3, and 4 1. Each progra feature set is a vector containing the aount of tie that each instruction would otherwise take to run by itself. These ties range fro 1 to 47 cycles. Feature weights are learned fro a training set of 1, progras, each, and results are plotted against a validation set of 2, progras each. Noralized error is calculated following: NoralizedError = pred iaca iaca 1 Section 4.1 extends this to real progras disassebled fro objdup. (1) Absolute value noralized error.6 Progra length (instructions) Figure 1: Noralized error for randoly generated progras using linear regression across progras of different lengths. where iaca is the nuber of cycles a gold standard, proprietary tool [1] predicts as progra runtie. By itself, linear regression perfors well: across all progra lengths tested, linear regression produces an average noralized error of only.158. The results in Figure 1 suggest an interesting trend: as progras increase in length, a basic linear regression works better and better. In soe ways, this is surprising. Althgouh the training set for each experient shown in Figure 1 is the sae size, the size of the space of potential progras that each odel tries to learn is very different: the space of progras of length 2 is roughly 15 1 larger than the space of progras of length 1 (there are approxiately 15 separate x86 instructions). This eans that for increasing progra lengths, each regression learns over a training set that is ore likely to be less and less representative of what it is trying to predict, i.e., for increasing progra lengths, we should expect ore error fro variance. Table 1 plots noralized average error for progras of length 1, 2, 3, and 4 as training set size increases. Note that after 1 saples, increases in training set size have alost no ipact on test set error. This suggests that the generalization error shown in Figure 1 is less a atter of under-representativeness of the training set and potentially ore a atter of bias in the underlying regression.
2 Prog S tr = 1 5 1k 2k 5k 1k 15k Len Table 1: Noralized error as training set size increases. 2.2 Kernelized regression The previous subsection argued that error fro linear regression was likely attributable to odel bias. Kernelizing the regression can reduce odel bias: whereas linear regression only captures linear dependencies between the features and the output of the odel, kernelizing can capture ore coplex relationships. We use the Kernel trick [6], replacing each original feature vector, x (i), with a (potentially) high-diensional feature vector, Φ(x (i) ) containing nonlinear functions of the input features, and then we solve linear regression using these high diensional vectors as our new feature vectors. Therefore, we solve: in. θ (θ Φ(x (i) ) y (i) ) 2 One issue in doing regression (especially nonlinear regressions) is that the solution to the regression proble can be a very coplex and non-soothe function, which can cause high variance. In order to overcoe this proble, we use Tikhonov regularization [5]. Specically, we add λ θ 2 2 to our cost function, which penalizes non-soothe predicted odels. (The intuition behind this is that if our predicted odel has saller coefficients then the odel will be ore sooth and we will have less variance). Therefore, we solve the following proble: in. θ (θ Φ(x (i) ) y (i) ) 2 + λ θ 2 2 Using the Kernel trick and the Representer Theore [4], we can show that the where: y predicted = θ Φ(x (i) ) = c = (K + λi ) 1 y c i K(x (i),x (new) ) Using these results, we expanded our linear regression to use: 1. a Gaussian kernel with paraeter σ (bandwidth) in which k(x (i),x ( j) ) = e x(i) x( j) 2 2 2σ 2 and 2. a polynoial kernel with paraeters d(degree) and c in which k(x (i),x ( j) ) = ((x (i) ) T x ( j) + c) d Absolute value noralized error Absolute value noralized error.5 Progra length (instructions) (a) Gaussian kernel regression Progra length (instructions) (b) Polynoial kernel regression Figure 2: Noralized error for Gaussian and polynoial kernels on fixed size progras. Figure 2 presents the results of the Gaussian (σ 2 =1) and polynoial kernelized regresssion (d=2, c=1). Both approaches showed only ild iproveents copared to linear regression: on average, 3% for the Gaussian kernel and 2% for the polynoial. There are several possible reasons that these kernels only show odest iproveents. Section 3.2 explores one of the ost basic of these: changing the apping fro x86 progra text to features. 3 Variable length progras The previous section discussed predicting runties for a set of progras of fixed length. However, progras of interest can have varying nubers of instructions: a superoptiizing copiler is priarily concerned with the question What is the fastest equivalent progra of any nuber of instructions? not What is the fastest equivalent progra of ten instructions? A patchwork approach that trains separate odels over the entire space of progra lengths following the ethodology outlined in Section 2.1 is feasible for our target application. However, fro a achine learning standpoint,
3 sub_progra_vec = []; runtie_estiate = instruction inst; for inst = progra.first_ins until inst = prgra.last_inst: runtie_estiate += increent_tie(sub_progra_vec, inst) sub_progra_vec.append(inst) end Figure 3: Pseudo code for generative runtie prediction algorith. it is relatively uninteresting. This section instead explores a different approach. It attepts to build a generative odel of progra runtie. Instead of predicting the runtie of a fixed length progra, the generative odel estiates the length of a subset of a progra and iteratively updates this estiate as additional instructions are added to the end of the progra. Figure 3 shows high-level pseudocode for this algorith. In essence, a generative odel predicts the increental change a single additional instruction akes to progra runtie. A good generative odel therefore addresses the challenge of progras of different lengths: one can predict the runtie of an arbitrarily-length progra by progressively updating a runtie estiate for each instruction in the progra. Section 3.1 presents a naive generative odel. Section 3.2 analyzes the deficiencies of this naive odel and uses the to otivate ore coplicated feature selection discussed. Section 3.3 uses these features to build ore coplicated generative odels and discusses their perforance. 3.1 Naive generative odel This section describes and shows results on progras fro ipleenting the siplest possible generative odel, which we call the naive generative odel. The naive generative odel decoposes a progra into its individual coponent instructions. It then assigns each instruction a weight that is the (precoputed) average of the tie it takes to run that instruction on all sets of valid inputs (e.g., registers and iediates). Finally, the naive generative odel calculates the overall progra runtie as the su of each of its instructions weights. Figure 4 shows the perforance of this odel for 1 progras coposed of a single instruction, four instructions, eight instructions, and sixteen instructions, each. The vertical axis of each subfigure shows each progra s runtie (in cycles) predicted fro a gold standard proprietary tool [1] and the horizontal axis shows the runtie predicted by the naive generative odel. Each subfigure shows the 45 o line corresponding to optial predictions: any point on this line is a correct prediction by the naive generative odel of progra runtie. Note that although each subfigure presents data over 1 3 progras, fewer than 1 points are distinguishable because any points overlap. This is particularly true for progras coposed of single instructions, which can achieve fewer predicted outcoes than larger progras coposed of any instructions. A least squares line (plotted in green) gives a sense of data spread for overlapping data. Gold standard prediction 2 15 Prog size = 1 Optial predictor slope = Naive linear prediction 35 8 Optial predictor slope = Optial predictor slope = Optial predictor slope =.48 7 Figure 4: Naive generative prediction error. The horizontal axis shows the nuber of cycles that the naive generative odel predicts; the vertical axis shows the nuber of cycles predicted using the gold standard IACA tool [1]. An optial predictor would (predicted cycles equals gold standard cycles) is labeled on each graph as a black line. The green line shows the results of fitting a generative regression through the data. As can be seen in Figure 4, as progra size increases, the naive generative odel perfors poorly: average error for single instruction progras is alost zero (as indicated by how closely the generative regression line atches the optial line); while the naive predictor for sixteen instruction progras, on average, has a isprediction error of alost 1%. This trend is to be expected. Iplicilty, the naive generative odel essentially trains across the space of single instruction progras. Therefore, when it is tested on a single instruction progra, it predicts well. However, when it is tested on progras with ore instructions (that therefore were not in its training set), it perfors poorly. The likely cause for this error is that the naive generative odel ignores processors potential parallelization and pipelining. Training a odel to independently predict runtie of individual instructions and suing the prediction for each does not capture this feature of processors. The following section explores appings that preserve register read-write conflicts. 3.2 Feature apping The degree that a progra can be parallelized partially depends on read-write dependencies on registers between instructions. For instance, consider the two, short progras shown in Figure 5. The seantics of each are siple. In the
4 Progra 1: ovl $x4, %rax addl $x4, %rbx, %rbx Progra 2: ovl $x4, %rax addl $x4, %rax, %rax Figure 5: Read-write dependencies first, the ovl instruction sets the value of register rax to a constant (x4); the addl instruction adds a constant (x4) to the contents of register rbx and puts this result in rbx. The second progra is siilar, except the addl instruction is over rax. In the first progra, the processor can run both the ovl and addl instructions in parallel. However, in the second, the addl instruction explicitly reads fro a register (rax) the ovl instruction writes to, and therefore the processor cannot run both in parallel. Therefore, although coposed of identical instructions, Progra 2 takes longer than Progra 1 to run. Both of the feature vectors described in the experients of Sections 3.1 and 2.1 did not track register dependencies: for both Progras 1 and 2, they produce identical feature vectors. To iprove results for the generative algorith, we updated our feature apping to include read-write dependencies. For each instruction in a progra, we parse the instruction s nae as well as each of the instruction s register arguents. Iportantly, registers with different naes can refer to parts of the sae physical register [2]. For instance, the nae al refers to the lower order bits of the sae physical register that the nae eax refers to. We track this aliasing as well as which portions of a register can be operated on concurrently. Given an instruction s nae, we index into a pre-copiled database that returns 1. A list of iplicit registers registers not specified by the arguents to the instruction that the instruction reads fro and writes to A list of which of its arguents it reads fro and writes to. For instance, the database ay specify that the instruction ovx a,b reads fro its first arguent (a) and writes to its second arguent (b), but does not operate on any iplicit registers. Following parsing and this database lookup, for every line in a progra, we can define its full register read and write set. Fro these read and write sets, we can easily infer register conflicts between a progra s instructions. 2 Instructions can also zero-extend registers, which writes zeros across the higher order bits of a physical register. 4 We have built a feature apping function, φ, paraeterized by a bandwidth paraeter τ, that odels these readwrite dependencies. More concretely, define a progra, P, as the ordered set I 1,I 2,I 3,...,I, where I i is the ith instruction in P. Then the feature vector associated with appending the instruction I +1 to this progra is C(I +1 τ ) 1{I +1 τ I +1 } C(I +2 τ ) 1{I +2 τ I +1 } φ(i +1,τ,P) =.. C(I ) 1{I i I +1 } C(I +1 ) where, C(.) is the nuber of cycles it takes to run a progra, and I j I k is true if the j th instruction of P writes to a register that the k th instruction of P reads or writes to. In this odel, the bandwidth paraeter, τ, specifies how any instructions to look back fro the current instruction for readwrite conflicts. Section 4.2 discusses the perforance of this feature creation approach, as any proposed solution should be fast enough to be useful in practice. 3.3 Generative odel In order to predict the increental runtie that each instruction adds to the total progra runtie, we have devised the following learning process. For progras of length 1, take each instruction, say I, and collect the features as explained in Section 3.2. Then run the progra with and without I, and use linear regression to learn the increental runtie that I adds to the progra fro the collected features. For a given progra with arbitrary length L, we add the predicted increental runtie of each instruction based on the collected features to predict the total runtie. Figure 6 shows the results. As the bandwidth increases the predictions becoes ore accurate. However, the difference for bandwidths greater than 5 is insignificant. Note that as we saw in Section 2.2, the Gaussian or polynoial kernel are alost the sae as linear regression so we only show the result for linear regression, here. 4 Conclusions and extensions This report exained using linear and kernelized regressions to predict progra runtie for x86 progras. For progras of fixed length we were able to predict progra runtie with 12% average error copared to a gold-standard tool. For variable lengthed progras, our generative odel perfors siilarly, and is flexible enough to predict progras of arbitrary size. The following subsections discuss additional experients we ran using our odel.
5 Absolute value noralized error bw=1 bw=2 bw=3 bw=4 bw=5 Absolute value noralized error bw=1 bw=2 bw=3 bw=4 bw=5. Progra length. Progra length Figure 6: Noralized error of generative odel with different bandwidth paraeter. Training set size is coposed of 8 progras with length 1. But, each test set has 2 progras of length 1, 2, 3, and 4. Figure 7: Noralized error of generative odel for disassebly of gcc using linear regression with a training set size 8 of and test set size of Real progras The dataset introduced in Section 2.1 and used throughout this report was coposed of rando sequences of instructions. This rando dataset atched the project sponsor s goals to intelligently search the space of all x86 progras. However, we were curious how well our odel predicted runties of copiler-generated progras, which have ore regular structure. To explore this question, we ran a disassebler (objdup) across progras copiled using gcc. Because the odel produced in the previous sections was for loopfree progras, we filtered jup instructions fro the disassebled code. This filtering liits our ability to draw strong conclusions about the odel s accuracy for real progras, but likely still produces ore realistic code segents than the previous randoly-generated dataset. Figure 7 shows the noralized error fro running the generative odel with linear regression on a disassebly of the gcc binary. Running the sae regressions on other progra types (e.g., progras that ake heavy use of IO, gui-based progras, etc.) produced siilar results. Average noralized error is approxiately 5% higher for this real data. 4.2 Feature collection runtie Our collaborator s target application requires not only predicting runtie accurately, but quickly as well. To ensure that the feature collection described in Section 3.2 does not becoe a prediction bottleneck, we bencharked the perforance of our unoptiized Python feature collector. Table2 shows these nubers collected on a 2.8GHz processor. With a bandwidth paraeter of 5, our unoptiized Python feature collector is able to produce features at a rate of approxiately 1k/second. Coupling these nubers with the tie required to perfor linear regression, the generative 5 bw= Instr/sec 154,52 133, ,97 97,38 Table 2: Feature vector construction tie versus bandwidth. odel described in Section 3.3 produces predictions orders of agnitude faster than issuing a syste call to execute the gold standard tool (65 predictions/s for generative linear regression odel on progras of 1 instructions vs 22 predictions/s for the gold-standard tool). References [1] Intel Architecture Code Analzyer. software.intel.co/en-us/articles/ intel-architecture-code-analyzer. [2] x86 Structure. X86#Structure. [3] E. Schkufza, R. Shara, and A. Aiken. Stochastic superoptiization. In Proceedings of the eighteenth international conference on Architectural support for prograing languages and operating systes, pages ACM, 213. [4] B. Schölkopf, R. Herbrich, and A. J. Sola. A generalized representer theore. In Coputational learning theory, pages Springer, 21. [5] A. N. Tikhonov and V. Y. Arsenin. Solutions of ill-posed probles [6] V. Vapnik. The nature of statistical learning theory. springer, 2.
Solving the Damage Localization Problem in Structural Health Monitoring Using Techniques in Pattern Classification
Solving the Daage Localization Proble in Structural Health Monitoring Using Techniques in Pattern Classification CS 9 Final Project Due Dec. 4, 007 Hae Young Noh, Allen Cheung, Daxia Ge Introduction Structural
More informationTensorFlow and Keras-based Convolutional Neural Network in CAT Image Recognition Ang LI 1,*, Yi-xiang LI 2 and Xue-hui LI 3
2017 2nd International Conference on Coputational Modeling, Siulation and Applied Matheatics (CMSAM 2017) ISBN: 978-1-60595-499-8 TensorFlow and Keras-based Convolutional Neural Network in CAT Iage Recognition
More informationEE 364B Convex Optimization An ADMM Solution to the Sparse Coding Problem. Sonia Bhaskar, Will Zou Final Project Spring 2011
EE 364B Convex Optiization An ADMM Solution to the Sparse Coding Proble Sonia Bhaskar, Will Zou Final Project Spring 20 I. INTRODUCTION For our project, we apply the ethod of the alternating direction
More informationImage Processing for fmri John Ashburner. Wellcome Trust Centre for Neuroimaging, 12 Queen Square, London, UK.
Iage Processing for fmri John Ashburner Wellcoe Trust Centre for Neuroiaging, 12 Queen Square, London, UK. Contents * Preliinaries * Rigid-Body and Affine Transforations * Optiisation and Objective Functions
More informationEvaluation of a multi-frame blind deconvolution algorithm using Cramér-Rao bounds
Evaluation of a ulti-frae blind deconvolution algorith using Craér-Rao bounds Charles C. Beckner, Jr. Air Force Research Laboratory, 3550 Aberdeen Ave SE, Kirtland AFB, New Mexico, USA 87117-5776 Charles
More informationI ve Seen Enough : Incrementally Improving Visualizations to Support Rapid Decision Making
I ve Seen Enough : Increentally Iproving Visualizations to Support Rapid Decision Making Sajjadur Rahan Marya Aliakbarpour 2 Ha Kyung Kong Eric Blais 3 Karrie Karahalios,4 Aditya Paraeswaran Ronitt Rubinfield
More informationTheoretical Analysis of Local Search and Simple Evolutionary Algorithms for the Generalized Travelling Salesperson Problem
Theoretical Analysis of Local Search and Siple Evolutionary Algoriths for the Generalized Travelling Salesperson Proble Mojgan Pourhassan ojgan.pourhassan@adelaide.edu.au Optiisation and Logistics, The
More informationIdentifying Converging Pairs of Nodes on a Budget
Identifying Converging Pairs of Nodes on a Budget Konstantina Lazaridou Departent of Inforatics Aristotle University, Thessaloniki, Greece konlaznik@csd.auth.gr Evaggelia Pitoura Coputer Science and Engineering
More informationEfficient Learning of Generalized Linear and Single Index Models with Isotonic Regression
Efficient Learning of Generalized Linear and Single Index Models with Isotonic Regression Sha M. Kakade Microsoft Research and Wharton, U Penn skakade@icrosoft.co Varun Kanade SEAS, Harvard University
More informationPerformance Analysis of RAID in Different Workload
Send Orders for Reprints to reprints@benthascience.ae 324 The Open Cybernetics & Systeics Journal, 2015, 9, 324-328 Perforance Analysis of RAID in Different Workload Open Access Zhang Dule *, Ji Xiaoyun,
More informationBrian Noguchi CS 229 (Fall 05) Project Final Writeup A Hierarchical Application of ICA-based Feature Extraction to Image Classification Brian Noguchi
A Hierarchical Application of ICA-based Feature Etraction to Iage Classification Introduction Iage classification poses one of the greatest challenges in the achine vision and achine learning counities.
More informationClustering. Cluster Analysis of Microarray Data. Microarray Data for Clustering. Data for Clustering
Clustering Cluster Analysis of Microarray Data 4/3/009 Copyright 009 Dan Nettleton Group obects that are siilar to one another together in a cluster. Separate obects that are dissiilar fro each other into
More informationGeo-activity Recommendations by using Improved Feature Combination
Geo-activity Recoendations by using Iproved Feature Cobination Masoud Sattari Middle East Technical University Ankara, Turkey e76326@ceng.etu.edu.tr Murat Manguoglu Middle East Technical University Ankara,
More informationDetection of Outliers and Reduction of their Undesirable Effects for Improving the Accuracy of K-means Clustering Algorithm
Detection of Outliers and Reduction of their Undesirable Effects for Iproving the Accuracy of K-eans Clustering Algorith Bahan Askari Departent of Coputer Science and Research Branch, Islaic Azad University,
More informationStructural Balance in Networks. An Optimizational Approach. Andrej Mrvar. Faculty of Social Sciences. University of Ljubljana. Kardeljeva pl.
Structural Balance in Networks An Optiizational Approach Andrej Mrvar Faculty of Social Sciences University of Ljubljana Kardeljeva pl. 5 61109 Ljubljana March 23 1994 Contents 1 Balanced and clusterable
More informationHomework 1. An Introduction to Neural Networks
Hoework An Introduction to Neural Networks -785: Introduction to Deep Learning Spring 09 OUT: January 4, 09 DUE: February 6, 09, :59 PM Start Here Collaboration policy: You are expected to coply with the
More informationA Hybrid Network Architecture for File Transfers
JOURNAL OF IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 6, NO., JANUARY 9 A Hybrid Network Architecture for File Transfers Xiuduan Fang, Meber, IEEE, Malathi Veeraraghavan, Senior Meber,
More informationBoosted Detection of Objects and Attributes
L M M Boosted Detection of Objects and Attributes Abstract We present a new fraework for detection of object and attributes in iages based on boosted cobination of priitive classifiers. The fraework directly
More informationA Novel Fast Constructive Algorithm for Neural Classifier
A Novel Fast Constructive Algorith for Neural Classifier Xudong Jiang Centre for Signal Processing, School of Electrical and Electronic Engineering Nanyang Technological University Nanyang Avenue, Singapore
More informationData pre-processing framework in SPM. Bogdan Draganski
Data pre-processing fraework in SPM Bogdan Draganski Outline Why do we need pre-processing? Overview Structural MRI pre-processing fmri pre-processing Why do we need pre-processing? What do we want? Reason
More informationCOMPUTER GENERATED HOLOGRAMS Optical Sciences 627 W.J. Dallas (Monday, August 23, 2004, 12:38 PM) PART III: CHAPTER ONE DIFFUSERS FOR CGH S
COPUTER GEERATED HOLOGRAS Optical Sciences 67 W.J. Dallas (onday, August 3, 004, 1:38 P) PART III: CHAPTER OE DIFFUSERS FOR CGH S Part III: Chapter One Page 1 of 8 Introduction Hologras for display purposes
More informationUtility-Based Resource Allocation for Mixed Traffic in Wireless Networks
IEEE IFOCO 2 International Workshop on Future edia etworks and IP-based TV Utility-Based Resource Allocation for ixed Traffic in Wireless etworks Li Chen, Bin Wang, Xiaohang Chen, Xin Zhang, and Dacheng
More informationThe optimization design of microphone array layout for wideband noise sources
PROCEEDINGS of the 22 nd International Congress on Acoustics Acoustic Array Systes: Paper ICA2016-903 The optiization design of icrophone array layout for wideband noise sources Pengxiao Teng (a), Jun
More informationImage Filter Using with Gaussian Curvature and Total Variation Model
IJECT Vo l. 7, Is s u e 3, Ju l y - Se p t 016 ISSN : 30-7109 (Online) ISSN : 30-9543 (Print) Iage Using with Gaussian Curvature and Total Variation Model 1 Deepak Kuar Gour, Sanjay Kuar Shara 1, Dept.
More informationEfficient Estimation of Inclusion Coefficient using HyperLogLog Sketches
Efficient Estiation of Inclusion Coefficient using HyperLogLog Sketches Azade Nazi, Bolin Ding, Vivek Narasayya, Surajit Chaudhuri Microsoft Research {aznazi, bolind, viveknar, surajitc}@icrosoft.co ABSTRACT
More informationA Trajectory Splitting Model for Efficient Spatio-Temporal Indexing
A Trajectory Splitting Model for Efficient Spatio-Teporal Indexing Slobodan Rasetic Jörg Sander Jaes Elding Mario A. Nasciento Departent of Coputing Science University of Alberta Edonton, Alberta, Canada
More informationFair Energy-Efficient Sensing Task Allocation in Participatory Sensing with Smartphones
Advance Access publication on 24 February 2017 The British Coputer Society 2017. All rights reserved. For Perissions, please eail: journals.perissions@oup.co doi:10.1093/cojnl/bxx015 Fair Energy-Efficient
More informationA Broadband Spectrum Sensing Algorithm in TDCS Based on ICoSaMP Reconstruction
MATEC Web of Conferences 73, 03073(08) https://doi.org/0.05/atecconf/087303073 SMIMA 08 A roadband Spectru Sensing Algorith in TDCS ased on I Reconstruction Liu Yang, Ren Qinghua, Xu ingzheng and Li Xiazhao
More informationResource Optimization for Web Service Composition
Resource Optiization for Web Coposition Xia Gao, Ravi Jain, Zulfikar Razan, Ulas Kozat Abstract coposition recently eerged as a costeffective way to quickly create new services within a network. Soe research
More informationGuillotine subdivisions approximate polygonal subdivisions: Part III { Faster polynomial-time approximation schemes for
Guillotine subdivisions approxiate polygonal subdivisions: Part III { Faster polynoial-tie approxiation schees for geoetric network optiization Joseph S. B. Mitchell y April 19, 1997; Last revision: May
More informationDesigning High Performance Web-Based Computing Services to Promote Telemedicine Database Management System
Designing High Perforance Web-Based Coputing Services to Proote Teleedicine Database Manageent Syste Isail Hababeh 1, Issa Khalil 2, and Abdallah Khreishah 3 1: Coputer Engineering & Inforation Technology,
More informationClosing The Performance Gap between Causal Consistency and Eventual Consistency
Closing The Perforance Gap between Causal Consistency and Eventual Consistency Jiaqing Du Călin Iorgulescu Aitabha Roy Willy Zwaenepoel EPFL ABSTRACT It is well known that causal consistency is ore expensive
More informationThe Internal Conflict of a Belief Function
The Internal Conflict of a Belief Function Johan Schubert Abstract In this paper we define and derive an internal conflict of a belief function We decopose the belief function in question into a set of
More informationFeature Selection to Relate Words and Images
The Open Inforation Systes Journal, 2009, 3, 9-13 9 Feature Selection to Relate Words and Iages Wei-Chao Lin 1 and Chih-Fong Tsai*,2 Open Access 1 Departent of Coputing, Engineering and Technology, University
More informationA Comparative Study of Two-phase Heuristic Approaches to General Job Shop Scheduling Problem
IE Vol. 7, No. 2, pp. 84-92, Septeber 2008. A Coparative Study of Two-phase Heuristic Approaches to General Job Shop Scheduling Proble Ji Ung Sun School of Industrial and Managent Engineering Hankuk University
More informationTHE rapid growth and continuous change of the real
IEEE TRANSACTIONS ON SERVICES COMPUTING, VOL. 8, NO. 1, JANUARY/FEBRUARY 2015 47 Designing High Perforance Web-Based Coputing Services to Proote Teleedicine Database Manageent Syste Isail Hababeh, Issa
More informationEnergy-Efficient Disk Replacement and File Placement Techniques for Mobile Systems with Hard Disks
Energy-Efficient Disk Replaceent and File Placeent Techniques for Mobile Systes with Hard Disks Young-Jin Ki School of Coputer Science & Engineering Seoul National University Seoul 151-742, KOREA youngjk@davinci.snu.ac.kr
More informationMAPPING THE DATA FLOW MODEL OF COMPUTATION INTO AN ENHANCED VON NEUMANN PROCESSOR * Peter M. Maurer
MAPPING THE DATA FLOW MODEL OF COMPUTATION INTO AN ENHANCED VON NEUMANN PROCESSOR * Peter M. Maurer Departent of Coputer Science and Engineering University of South Florida Tapa, FL 33620 Abstract -- The
More informationVisualization of Support Vector Machines with Unsupervised Learning
Visualization of Support Vector Machines with Unsupervised Learning Lutz Hael Departent of Coputer Science and Statistics University of Rhode Island, Kingston, RI 02881 lutz@inductive-reasoning.co Abstract
More informationCS 361 Meeting 8 9/24/18
CS 36 Meeting 8 9/4/8 Announceents. Hoework 3 due Friday. Review. The closure properties of regular languages provide a way to describe regular languages by building the out of sipler regular languages
More informationPreprocessing of fmri data (basic)
Preprocessing of fmri data (basic) Practical session SPM Course 2016, Zurich Andreea Diaconescu, Maya Schneebeli, Jakob Heinzle, Lars Kasper, and Jakob Sieerkus Translational Neuroodeling Unit (TNU) Institute
More informationELEVATION SURFACE INTERPOLATION OF POINT DATA USING DIFFERENT TECHNIQUES A GIS APPROACH
ELEVATION SURFACE INTERPOLATION OF POINT DATA USING DIFFERENT TECHNIQUES A GIS APPROACH Kulapraote Prathuchai Geoinforatics Center, Asian Institute of Technology, 58 Moo9, Klong Luang, Pathuthani, Thailand.
More informationComputer Aided Drafting, Design and Manufacturing Volume 26, Number 2, June 2016, Page 13
Coputer Aided Drafting, Design and Manufacturing Volue 26, uber 2, June 2016, Page 13 CADDM 3D reconstruction of coplex curved objects fro line drawings Sun Yanling, Dong Lijun Institute of Mechanical
More informationDefining and Surveying Wireless Link Virtualization and Wireless Network Virtualization
1 Defining and Surveying Wireless Link Virtualization and Wireless Network Virtualization Jonathan van de Belt, Haed Ahadi, and Linda E. Doyle The Centre for Future Networks and Counications - CONNECT,
More informationDiscrete Fourier Transform
Discrete Fourier Transfor This is the first tutorial in our ongoing series on tie series spectral analysis. In this entry, we will closely exaine the discrete Fourier transfor (aa DFT) and its inverse,
More informationData & Knowledge Engineering
Data & Knowledge Engineering 7 (211) 17 187 Contents lists available at ScienceDirect Data & Knowledge Engineering journal hoepage: www.elsevier.co/locate/datak An approxiate duplicate eliination in RFID
More information2013 IEEE Conference on Computer Vision and Pattern Recognition. Compressed Hashing
203 IEEE Conference on Coputer Vision and Pattern Recognition Copressed Hashing Yue Lin Rong Jin Deng Cai Shuicheng Yan Xuelong Li State Key Lab of CAD&CG, College of Coputer Science, Zhejiang University,
More informationA Learning Framework for Nearest Neighbor Search
A Learning Fraework for Nearest Neighbor Search Lawrence Cayton Departent of Coputer Science University of California, San Diego lcayton@cs.ucsd.edu Sanjoy Dasgupta Departent of Coputer Science University
More informationPlanet Scale Software Updates
Planet Scale Software Updates Christos Gkantsidis 1, Thoas Karagiannis 2, Pablo Rodriguez 1, and Milan Vojnović 1 1 Microsoft Research 2 UC Riverside Cabridge, UK Riverside, CA, USA {chrisgk,pablo,ilanv}@icrosoft.co
More informationAdaptive Holistic Scheduling for In-Network Sensor Query Processing
Adaptive Holistic Scheduling for In-Network Sensor Query Processing Hejun Wu and Qiong Luo Departent of Coputer Science and Engineering Hong Kong University of Science & Technology Clear Water Bay, Kowloon,
More informationError estimation of bathymetric grid models derived from historic and contemporary datasets 1 Introduction
Error estiation of bathyetric grid odels derived fro historic and conteporary datasets Martin Jakobsson, Brian Calder, Larry Mayer and Andy Arstrong Center for Coastal and Ocean Mapping, University of
More informationCarving Differential Unit Test Cases from System Test Cases
Carving Differential Unit Test Cases fro Syste Test Cases Sebastian Elbau, Hui Nee Chin, Matthew B. Dwyer, Jonathan Dokulil Departent of Coputer Science and Engineering University of Nebraska - Lincoln
More informationOptimized stereo reconstruction of free-form space curves based on a nonuniform rational B-spline model
1746 J. Opt. Soc. A. A/ Vol. 22, No. 9/ Septeber 2005 Y. J. Xiao and Y. F. Li Optiized stereo reconstruction of free-for space curves based on a nonunifor rational B-spline odel Yi Jun Xiao Departent of
More informationA Study of the Relationship Between Support Vector Machine and Gabriel Graph
A Study of the Relationship Between Support Vector Machine and Gabriel Graph Wan Zhang and Irwin King {wzhang, king}@cse.cuhk.edu.hk Departent of Coputer Science & Engineering The Chinese University of
More informationNON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH
NON-RIGID OBJECT TRACKING: A PREDICTIVE VECTORIAL MODEL APPROACH V. Atienza; J.M. Valiente and G. Andreu Departaento de Ingeniería de Sisteas, Coputadores y Autoática Universidad Politécnica de Valencia.
More informationRule Extraction using Artificial Neural Networks
Rule Extraction using Artificial Neural Networks S. M. Karuzzaan 1 Ahed Ryadh Hasan 2 Abstract Artificial neural networks have been successfully applied to a variety of business application probles involving
More informationA Measurement-Based Model for Parallel Real-Time Tasks
A Measureent-Based Model for Parallel Real-Tie Tasks Kunal Agrawal 1 Washington University in St. Louis St. Louis, MO, USA kunal@wustl.edu https://orcid.org/0000-0001-5882-6647 Sanjoy Baruah 2 Washington
More informationSpectral Clustering with Two Views
Spectral Clustering with Two Views Virginia R. de Sa Departent of Cognitive Science, 55 University of California, San Diego 95 Gilan Dr. La Jolla, CA 993-55 desa@ucsd.edu Abstract In this paper we develop
More informationApproximate String Matching with Reduced Alphabet
Approxiate String Matching with Reduced Alphabet Leena Salela 1 and Jora Tarhio 2 1 University of Helsinki, Departent of Coputer Science leena.salela@cs.helsinki.fi 2 Aalto University Deptartent of Coputer
More informationResearch on a Kind of QoS-Sensitive Semantic Web Services Composition Method Based on Genetic Algorithm
roceedings of the 7th International Conference on Innovation & Manageent 893 Research on a Kind of QoS-Sensitive Seantic Web Services Coposition Method Based on Genetic Algorith Cao Hongjiang, Nie Guihua,
More informationAvailable online at Procedia Computer Science 9 (2012 ) International Conference on Computational Science, ICCS 2012
Available online at www.sciencedirect.co Procedia Coputer Science 9 (2012 ) 1363 1370 Abstract International Conference on Coputational Science, ICCS 2012 Reachability Analysis of Concurrent Boolean Progras
More informationSmarter Balanced Assessment Consortium Claims, Targets, and Standard Alignment for Math
Sarter Balanced Assessent Consortiu s, s, Stard Alignent for Math The Sarter Balanced Assessent Consortiu (SBAC) has created a hierarchy coprised of clais targets that together can be used to ake stateents
More informationAdaptive Parameter Estimation Based Congestion Avoidance Strategy for DTN
Proceedings of the nd International onference on oputer Science and Electronics Engineering (ISEE 3) Adaptive Paraeter Estiation Based ongestion Avoidance Strategy for DTN Qicai Yang, Futong Qin, Jianquan
More informationλ-harmonious Graph Colouring Lauren DeDieu
λ-haronious Graph Colouring Lauren DeDieu June 12, 2012 ABSTRACT In 198, Hopcroft and Krishnaoorthy defined a new type of graph colouring called haronious colouring. Haronious colouring is a proper vertex
More informationCOLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL
COLOR HISTOGRAM AND DISCRETE COSINE TRANSFORM FOR COLOR IMAGE RETRIEVAL 1 Te-Wei Chiang ( 蔣德威 ), 2 Tienwei Tsai ( 蔡殿偉 ), 3 Jeng-Ping Lin ( 林正平 ) 1 Dept. of Accounting Inforation Systes, Chilee Institute
More informationClassification and Prediction of Motion Trajectories using Spatiotemporal Approximations
Classification and Prediction of Motion Trajectories using Spatioteporal Approxiations Andrew Naftel & Shehzad Khalid School of Inforatics University of Manchester Manchester, M6 1QD a.naftel@anchester.ac.uk;
More informationAGV PATH PLANNING BASED ON SMOOTHING A* ALGORITHM
International Journal of Software Engineering & Applications (IJSEA), Vol.6, No.5, Septeber 205 AGV PATH PLANNING BASED ON SMOOTHING A* ALGORITHM Xie Yang and Cheng Wushan College of Mechanical Engineering,
More informationGromov-Hausdorff Distance Between Metric Graphs
Groov-Hausdorff Distance Between Metric Graphs Jiwon Choi St Mark s School January, 019 Abstract In this paper we study the Groov-Hausdorff distance between two etric graphs We copute the precise value
More informationA Learning Framework for Nearest Neighbor Search
A Learning Fraework for Nearest Neighbor Search Lawrence Cayton Departent of Coputer Science University of California, San Diego lcayton@cs.ucsd.edu Sanjoy Dasgupta Departent of Coputer Science University
More informationGAN-OPC: Mask Optimization with Lithography-guided Generative Adversarial Nets
GAN-OPC: Mask Optiization with Lithography-guided Generative Adversarial Nets Haoyu Yang hyyang@cse.cuhk.edu.hk Shuhe Li shli@cse.cuhk.edu.hk Yuzhe Ma yza@cse.cuhk.edu.hk ABSTRACT Bei Yu byu@cse.cuhk.edu.hk
More informationOPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS
Key words SOA, optial, coplex service, coposition, Quality of Service Piotr RYGIELSKI*, Paweł ŚWIĄTEK* OPTIMAL COMPLEX SERVICES COMPOSITION IN SOA SYSTEMS One of the ost iportant tasks in service oriented
More informationAn Ad Hoc Adaptive Hashing Technique for Non-Uniformly Distributed IP Address Lookup in Computer Networks
An Ad Hoc Adaptive Hashing Technique for Non-Uniforly Distributed IP Address Lookup in Coputer Networks Christopher Martinez Departent of Electrical and Coputer Engineering The University of Texas at San
More informationRelief shape inheritance and graphical editor for the landscape design
Relief shape inheritance and graphical editor for the landscape design Egor A. Yusov Vadi E. Turlapov Nizhny Novgorod State University after N. I. Lobachevsky Nizhny Novgorod Russia yusov_egor@ail.ru vadi.turlapov@cs.vk.unn.ru
More informationQuerying Uncertain Data with Aggregate Constraints
Querying Uncertain Data with Aggregate Constraints Mohan Yang 1,2, Haixun Wang 1, Haiquan Chen 3, Wei-Shinn Ku 3 1 Microsoft Research Asia, Beijing, China 2 Shanghai Jiao Tong University, Shanghai, China
More informationJoint Measurement- and Traffic Descriptor-based Admission Control at Real-Time Traffic Aggregation Points
Joint Measureent- and Traffic Descriptor-based Adission Control at Real-Tie Traffic Aggregation Points Stylianos Georgoulas, Panos Triintzios and George Pavlou Centre for Counication Systes Research, University
More informationEFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B.
EFFICIENT VIDEO SEARCH USING IMAGE QUERIES A. Araujo1, M. Makar2, V. Chandrasekhar3, D. Chen1, S. Tsai1, H. Chen1, R. Angst1 and B. Girod1 1 Stanford University, USA 2 Qualco Inc., USA ABSTRACT We study
More informationA Low-Cost Multi-Failure Resilient Replication Scheme for High Data Availability in Cloud Storage
216 IEEE 23rd International Conference on High Perforance Coputing A Low-Cost Multi-Failure Resilient Replication Schee for High Data Availability in Cloud Storage Jinwei Liu* and Haiying Shen *Departent
More informationIn the Proceedings of the 7 th ACM International Conference on Supercomputing,
In the Proceedings of the 7 th ACM International Conference on Supercoputing, Tokyo, Japan, July 20{22, 1993, pp. 77{86. c1993 ACM (see notice below). Speculative Execution and Branch Prediction on Parallel
More informationPERFORMANCE MEASURES FOR INTERNET SERVER BY USING M/M/m QUEUEING MODEL
IJRET: International Journal of Research in Engineering and Technology ISSN: 239-63 PERFORMANCE MEASURES FOR INTERNET SERVER BY USING M/M/ QUEUEING MODEL Raghunath Y. T. N. V, A. S. Sravani 2 Assistant
More informationMedical Biophysics 302E/335G/ st1-07 page 1
Medical Biophysics 302E/335G/500 20070109 st1-07 page 1 STEREOLOGICAL METHODS - CONCEPTS Upon copletion of this lesson, the student should be able to: -define the ter stereology -distinguish between quantitative
More informationOn the Computation and Application of Prototype Point Patterns
On the Coputation and Application of Prototype Point Patterns Katherine E. Tranbarger Freier 1 and Frederic Paik Schoenberg 2 Abstract This work addresses coputational probles related to the ipleentation
More informationAn Efficient Approach for Content Delivery in Overlay Networks
An Efficient Approach for Content Delivery in Overlay Networks Mohaad Malli, Chadi Barakat, Walid Dabbous Projet Planète, INRIA-Sophia Antipolis, France E-ail:{alli, cbarakat, dabbous}@sophia.inria.fr
More informationIN many interactive multimedia streaming applications, such
840 IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 8, NO. 5, MAY 06 A Geoetric Approach to Server Selection for Interactive Video Streaing Yaochen Hu, Student Meber, IEEE, DiNiu, Meber, IEEE, and Zongpeng Li, Senior
More informationA CRYPTANALYTIC ATTACK ON RC4 STREAM CIPHER
A CRYPTANALYTIC ATTACK ON RC4 STREAM CIPHER VIOLETA TOMAŠEVIĆ, SLOBODAN BOJANIĆ 2 and OCTAVIO NIETO-TALADRIZ 2 The Mihajlo Pupin Institute, Volgina 5, 000 Belgrade, SERBIA AND MONTENEGRO 2 Technical University
More informationFeature Based Registration for Panoramic Image Generation
IJCSI International Journal of Coputer Science Issues, Vol. 10, Issue 6, No, Noveber 013 www.ijcsi.org 13 Feature Based Registration for Panoraic Iage Generation Kawther Abbas Sallal 1, Abdul-Mone Saleh
More informationSecure Wireless Multihop Transmissions by Intentional Collisions with Noise Wireless Signals
Int'l Conf. Wireless etworks ICW'16 51 Secure Wireless Multihop Transissions by Intentional Collisions with oise Wireless Signals Isau Shiada 1 and Hiroaki Higaki 1 1 Tokyo Denki University, Japan Abstract
More informationGrading Results Total 100
University of California, Berkeley College of Engineering Departent of Electrical Engineering and Coputer Sciences Fall 2003 Instructor: Dave Patterson 2003-11-19 v1.9 CS 152 Exa #2 Solutions Personal
More informationA Novel Approach to Fractal Dimension based Fingerprint Recognition System
Volue: 03 Issue: 04 Apr-2016 www.irjet.net p-issn: 2395-0072 A Novel Approach to Fractal Diension based Fingerprint Recognition Syste Chiteranjan Sahu 1, Vinay Jain 2 1M.E. Scholar, Dept. of Electronics
More informationScheduling Parallel Real-Time Recurrent Tasks on Multicore Platforms
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL., NO., NOV 27 Scheduling Parallel Real-Tie Recurrent Tasks on Multicore Platfors Risat Pathan, Petros Voudouris, and Per Stenströ Abstract We
More informationDesign Optimization of Mixed Time/Event-Triggered Distributed Embedded Systems
Design Optiization of Mixed Tie/Event-Triggered Distributed Ebedded Systes Traian Pop, Petru Eles, Zebo Peng Dept. of Coputer and Inforation Science, Linköping University {trapo, petel, zebpe}@ida.liu.se
More informationOn computing global similarity in images
On coputing global siilarity in iages S Ravela and R Manatha Coputer Science Departent University of Massachusetts, Aherst, MA 01003 Eail: ravela,anatha @csuassedu Abstract The retrieval of iages based
More informationPOSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION. Junjun Jiang, Ruimin Hu, Zhen Han, Tao Lu, and Kebin Huang
IEEE International Conference on ultiedia and Expo POSITION-PATCH BASED FACE HALLUCINATION VIA LOCALITY-CONSTRAINED REPRESENTATION Junjun Jiang, Ruiin Hu, Zhen Han, Tao Lu, and Kebin Huang National Engineering
More informationModeling Parallel Applications Performance on Heterogeneous Systems
Modeling Parallel Applications Perforance on Heterogeneous Systes Jaeela Al-Jaroodi, Nader Mohaed, Hong Jiang and David Swanson Departent of Coputer Science and Engineering University of Nebraska Lincoln
More informationPrivacy-preserving String-Matching With PRAM Algorithms
Privacy-preserving String-Matching With PRAM Algoriths Report in MTAT.07.022 Research Seinar in Cryptography, Fall 2014 Author: Sander Sii Supervisor: Peeter Laud Deceber 14, 2014 Abstract In this report,
More informationAutomatic Die Inspection for Post-sawing LED Wafers
Proceedings of the 009 IEEE International Conference on Systes, Man, and Cybernetics San Antonio, TX, USA - ctober 009 Autoatic Die Inspection for Post-sawing ED Wafers Chuan-Yu Chang 1, Chun-Hsi i, Yung-Chi
More informationHeterogeneous Radial Basis Function Networks
Proceedings of the International Conference on Neural Networks (ICNN ), vol. 2, pp. 23-2, June. Heterogeneous Radial Basis Function Networks D. Randall Wilson, Tony R. Martinez e-ail: randy@axon.cs.byu.edu,
More informationINSERTION SORT is O(n log n)
INSERTION SORT is On log n) Michael A. Bender Martín Farach-Colton Miguel A. Mosteiro Abstract Traditional INSERTION SORT runs in On 2 ) tie because each insertion takes On) tie. When people run INSERTION
More informationQUERY ROUTING OPTIMIZATION IN SENSOR COMMUNICATION NETWORKS
QUERY ROUTING OPTIMIZATION IN SENSOR COMMUNICATION NETWORKS Guofei Jiang and George Cybenko Institute for Security Technology Studies and Thayer School of Engineering Dartouth College, Hanover NH 03755
More informationPreprocessing I: Within Subject John Ashburner
Preprocessing I: Within Subject John Ashburner Pre-processing Overview Statistics or whatever fmri tie-series Anatoical MRI Teplate Soothed Estiate Spatial Nor Motion Correct Sooth Coregister 11 21 31
More informationMapping Data in Peer-to-Peer Systems: Semantics and Algorithmic Issues
Mapping Data in Peer-to-Peer Systes: Seantics and Algorithic Issues Anastasios Keentsietsidis Marcelo Arenas Renée J. Miller Departent of Coputer Science University of Toronto {tasos,arenas,iller}@cs.toronto.edu
More information