CPU Execution Time Research Articles

State-Machine and Deferred-Update Replication: Analysis and Comparison

In the paper, we analyze and experimentally compare two popular replication schemes relying on atomic broadcast: state machine replication (SMR) and deferred update replication (DUR). We estimate the lower bounds on the time of executing requests by the SMR and DUR systems running on multi-core servers. We also consider variants of systems that can process read-only requests with a lower overhead. In the analysis of DUR, we consider conflict patterns. We then formally show the scalability of SMR and DUR, which reflects the capacity of systems to effectively utilize an increasing number of processor cores. Next, we compare SMR and DUR experimentally under different levels of contention, using several benchmarks. We show throughput, abort rate (in DUR), and network congestion. The key results of our work are that neither system is superior in all cases, and that the theoretical and experimental results are heavily influenced by the dominance of either the CPU execution time or atomic broadcast time. We therefore propose to combine both replication schemes and gain the best of both worlds.

Fast algorithm for hybrid region‐based active contours optimisation

Active contours are usually based on the optimisation of energy functionals that are built to attract the curve towards the objects' boundaries. This study describes a hybrid region-based active contours technique that uses global means to define the global fitting energy and local means and variances to define the local fitting energy. The optimisation of the functional is performed by applying a sweeping-principle algorithm, which avoids solving any partial differential equation and removes the need for any stability conditions. Furthermore, sweeping-principle algorithm is not based on the computation of derivatives, which allows using a binary level set function during the minimisation process instead of the signed distance function, consequently this removes the need for the distance regularisation term, avoiding its subtle side effects and speeding up the optimisation process. Successful and accurate segmentation results are obtained on synthetic and real images with a significant gain in the CPU execution time when compared with the minimisation via the commonly used gradient descent method.

Effects of Multicore Distributed Memory Systems on Parallel Processing Applications

Complex problems need a longtime to be solved, with low efficiency and performance. Hence, to overcome these drawbacks, the approach of breaking the problem into independent parts and treating each part individually in the way that each processing element can execute its part of the problem simultaneously with the others. The systems that contain many computing elements combined. Parallel processing (PP) is divided into three types; shared, distributed, and hybrid memory systems are usually adopted. The aim of this research is to point out the effects of multicore distributed memory systems on PP applications that can reduce the total execution time of the programs. In this work, distributed- and shared-memory systems addressed depends on client/servers principles. However, to get the exact evaluation of our aim, just one client and one server have been depended. The algorithm used here is capable of calculating: The started, consumed, and terminated for CPU and total execution times, CPU usage of servers, and CPU and Total execution times for the client. The results compared with previous works depending on distributed memory systems, to overcome the previous drawbacks taking in the consideration the effects of multi-core processor. All of these algorithms are implemented using Java Language.

Population Based Metaheuristic Algorithm Approach for Analysis of Multi-Item Multi-Period Procurement Lot Sizing Problem

This research study focuses on the optimization of multi-item multi-period procurement lot sizing problem for inventory management. Mathematical model is developed which considers different practical constraints like storage space and budget. The aim is to find optimum order quantities of the product so that total cost of inventory is minimized. The NP-hard mathematical model is solved by adopting a novel ant colony optimization approach. Due to lack of benchmark method specified in the literature to assess the performance of the above approach, another metaheuristic based program of genetic algorithm is also employed to solve the problem. The parameters of genetic algorithm model are calibrated using Taguchi method of experiments. The performance of both algorithms is compared using ANOVA analysis with the real time data collected from a valve manufacturing company. It is verified that two methods have not shown any significant difference as far as objective function value is considered. But genetic algorithm is far better than the ACO method when compared on the basis of CPU execution time.

A comparative study of evolutionary algorithms and adapting control parameters for estimating the parameters of a single-diode photovoltaic module's model

This paper proposes different evolutionary algorithms, such as differential evolution and electromagnetism-like algorithms, to extract the five parameters of a single-diode photovoltaic module's model. Hybrid evolutionary algorithms are proposed with integrated and adaptive mutation per iteration schemes. In addition, a new formula to adjust the mutation scaling factor and crossover rate for each generation is proposed. Analyses are performed based on experimental data points under different weather conditions to explain the robustness and reliability of the proposed methods. Results show that the proposed hybrid algorithms, namely, evolutionary algorithm with integrated mutation per iteration and evolutionary algorithm with adaptive mutation per iteration, exhibit better performance than electromagnetism-like algorithm and other methods in terms of accuracy, CPU execution time, and convergence. The proposed hybrid algorithms offer a root mean square error, mean bias error, coefficient of determination and CPU execution time around 0.062, 0.006 and 0.992, and less than 20 s respectively. Furthermore, the feasibility of the proposed methods is validated by comparing the obtained results with those of other methods under various statistical errors. As a conclusion, the proposed hybrid algorithms offer root mean square error and mean bias error less than other methods by 14% at least.

A Simulation-based Approach to Optimize the Execution Time and Minimization of Average Waiting Time Using Queuing Model in Cloud Computing Environment

Cloud computing is the emerging domain in academia and IT Industry. It is a business framework for delivering the services and computing power on-demand basis. Cloud users have to pay the service providers based on their usage. For enterprises, cloud computing is the worthy of consideration and they try to build business systems with lower costs, higher profits and quality-of-service. Considering cost optmization, service provider may initially try to use less number of CPU cores and data centers. For that reason, this paper deals with CloudSim simulation tool which has been utilized for evaluating the number of CPU cores and execution time. Minimization of waiting time is also a considerable issue. When a large number of jobs are requested, they have to wait for getting allocated to the servers which in turn may increase the queue length and also waiting time. This paper also deals with queuing model with multi-server and finite capacity to reduce the waiting time and queue length.

Exploiting Structure and Variable Dependency Modeling in Block-based Compressed Sensing Image Reconstruction in the Presence of Non-linear Mixtures

With the introduction of compressed sensing (CS) theory, investigation into exploiting sparseness and optimizing compressive sensing performance has ensued. Compressed sensing is highly applicable to images, which naturally have sparse representations. Improvements in the area of image denoising have resulted from the combination of highly-directional transforms with shrinkage and thresholding techniques along with imposition of a model to account for statistical properties of images. Using this approach, statistical modeling of dependencies in the transform domain is incorporated into high-performance and efficient state-of-the-art CS image reconstruction algorithms with highly-directional transforms incorporating redundancy and bivariate shrinkage and thresholding to further refine image reconstruction performance improvements. Additionally, hierarchical structural dependency modeling is incorporated to account for parent–child coefficient relationships. These techniques exploit hierarchical structure and multiscale subbands of frequencies and orientation, exploiting dependencies across and within scales. Additionally, these techniques are incorporated with minimal additional CPU execution time into block-based CS (BCS) algorithms, which are known for their efficient and fast computation time. Experimental results show increased refinements of image reconstruction performance over current state-of-the-art image reconstruction algorithms, particularly at the higher CS ratios (lower sampling rates) of interest in compressed sensing. © 2015 Society for Imaging Science and Technology. [DOI: 10.2352/J.ImagingSci.Technol.2015.59.6.060406]

Exploiting Structure and Variable Dependency Modeling in Block-based Compressed Sensing Image Reconstruction in the Presence of Non-linear Mixtures

With the introduction of compressed sensing (CS) theory, investigation into exploiting sparseness and optimizing compressive sensing performance has ensued. Compressed sensing is highly applicable to images, which naturally have sparse representations. Improvements in the area of image denoising have resulted from the combination of highly-directional transforms with shrinkage and thresholding techniques along with imposition of a model to account for statistical properties of images. Using this approach, statistical modeling of dependencies in the transform domain is incorporated into high-performance and efficient state-of-the-art CS image reconstruction algorithms with highly-directional transforms incorporating redundancy and bivariate shrinkage and thresholding to further refine image reconstruction performance improvements. Additionally, hierarchical structural dependency modeling is incorporated to account for parent‐child coefficient relationships. These techniques exploit hierarchical structure and multiscale subbands of frequencies and orientation, exploiting dependencies across and within scales. Additionally, these techniques are incorporated with minimal additional CPU execution time into block-based CS (BCS) algorithms, which are known for their efficient and fast computation time. Experimental results show increased refinements of image reconstruction performance over current state-of-the-art image reconstruction algorithms, particularly at the higher CS ratios (lower sampling rates) of interest in compressed sensing.

A full-spectrum k-distribution look-up table for radiative transfer in nonhomogeneous gaseous media

A full-spectrum k-distribution (FSK) look-up table has been constructed for gas mixtures within a certain range of thermodynamic states for three species, i.e., CO2, H2O and CO. The k-distribution of a mixture is assembled directly from the summation of the linear absorption coefficients of three species. The systematic approach to generate the table, including the generation of the pressure-based absorption coefficient and the generation of the k-distribution, is discussed. To efficiently obtain accurate k-values for arbitrary thermodynamic states from tabulated values, a 6-D linear interpolation method is employed. A large number of radiative heat transfer calculations have been carried out to test the accuracy of the FSK look-up table. Results show that, using the FSK look-up table can provide excellent accuracy compared to the exact results. Without the time-consuming process of assembling k-distribution from individual species plus mixing, using the FSK look-up table can save considerable computational cost. To evaluate the accuracy as well as the efficiency of the FSK look-up table, radiative heat transfer via a scaled Sandia D Flame is calculated to compare the CPU execution time using the FSK method based on the narrow-band database, correlations, and the look-up table. Results show that the FSK look-up table can provide a computationally cheap alternative without much sacrifice in accuracy.

Extraction of photovoltaic module model’s parameters using an improved hybrid differential evolution/electromagnetism-like algorithm

In this paper, an improved differential evolution with adaptive mutation per iteration algorithm (DEAM) is proposed for extracting PV module’s model parameters. DEAM utilizes the attraction–repulsion concept which is used in the electromagnetism to boost the mutation operation of the original differential evolution (DE). Furthermore, a new formula to adjust the mutation scaling factor and crossover rate for each generation is proposed. The proposed method has been validated by experimental data and other previous methods. The results of the proposed method show a high agreement between the experimental and simulated I–V characteristics. The average root mean square error, mean bias error, coefficient of determination and CPU-execution time of the proposed method are 1.744%, 0.158%, 99.21% and 18.5975s respectively. According to the results, the proposed method offers better performance than other methods in terms of accuracy, CPU-execution time and convergence.

CUDA Approach for Meshless Local Petrov–Galerkin Method

In this paper, a strategy to parallelize the meshless local Petrov–Galerkin (MLPG) method is developed. It is executed in a high parallel architecture, the well known graphics processing unit. The MLPG algorithm has many variations depending on which combination of trial and test functions is used. Two types of interpolation schemes are explored in this paper to approximate the trial functions and a Heaviside step function is used as test function. The first scheme approximates the trial function through a moving least squares interpolation, and the second interpolates using the radial point interpolation method with polynomial reproduction (RPIMp). To compare these two approaches, a simple electromagnetic problem is solved, and the number of nodes in the domain is increased while the time to assemble the system of equations is obtained. Results shows that with the parallel version of the algorithm it is possible to achieve an execution time 20 times smaller than the CPU execution time, for the MLPG using RPIMp versions of the method.

Comparative Analysis of Data Mining Classification Algorithms in Type-2 Diabetes Prediction Data Using WEKA Approach

The goal of this paper discusses about different types of data mining classification algorithms accuracies that are widely used to extract significant knowledge from huge amounts of data. Here illustrate 20 classifications of supervised data mining algorithms base on type-2 diabetes disease dataset perspective to Bangladeshi populations. In this paper we compare 20 classification algorithms by measuring accuracies, speed and robustness of those algorithms using WEKA toolkit version 3.6.5. Accuracies of classification algorithms are measured in 3 cases like Total Training data set, 10 fold Cross Validation and Percentage Split (66% taken). Speed (CPU Execution Time) and error rate also measured as like as accuracy. Firstly checked top perform algorithms that have best outcome for different cases and then ranked top outcomes algorithms. Finally ranked best 5 algorithms among 20 algorithms based on their accuracies.

PVT: an efficient computational procedure to speed up next-generation sequence analysis.

BackgroundHigh-throughput Next-Generation Sequencing (NGS) techniques are advancing genomics and molecular biology research. This technology generates substantially large data which puts up a major challenge to the scientists for an efficient, cost and time effective solution to analyse such data. Further, for the different types of NGS data, there are certain common challenging steps involved in analysing those data. Spliced alignment is one such fundamental step in NGS data analysis which is extremely computational intensive as well as time consuming. There exists serious problem even with the most widely used spliced alignment tools. TopHat is one such widely used spliced alignment tools which although supports multithreading, does not efficiently utilize computational resources in terms of CPU utilization and memory. Here we have introduced PVT (Pipelined Version of TopHat) where we take up a modular approach by breaking TopHat’s serial execution into a pipeline of multiple stages, thereby increasing the degree of parallelization and computational resource utilization. Thus we address the discrepancies in TopHat so as to analyze large NGS data efficiently.ResultsWe analysed the SRA dataset (SRX026839 and SRX026838) consisting of single end reads and SRA data SRR1027730 consisting of paired-end reads. We used TopHat v2.0.8 to analyse these datasets and noted the CPU usage, memory footprint and execution time during spliced alignment. With this basic information, we designed PVT, a pipelined version of TopHat that removes the redundant computational steps during ‘spliced alignment’ and breaks the job into a pipeline of multiple stages (each comprising of different step(s)) to improve its resource utilization, thus reducing the execution time.ConclusionsPVT provides an improvement over TopHat for spliced alignment of NGS data analysis. PVT thus resulted in the reduction of the execution time to ~23% for the single end read dataset. Further, PVT designed for paired end reads showed an improved performance of ~41% over TopHat (for the chosen data) with respect to execution time. Moreover we propose PVT-Cloud which implements PVT pipeline in cloud computing system.

Developing the Best Scheduling Algorithm From Existing Algorithms for Real Time Operating Systems

This paper is about the selection of the scheduling algorithm for the real-time system. In this paper, we compared different scheduling algorithms and from this comparison, we get another algorithm which is good in performance as compared to the existing one. First, we compare ACO and EDF but both have some merits and demerits. ACO is not good when the system is underloaded, preemptive and single processor while the result of ACO is good in overloaded condition with a lot of execution time. From the comparison of ACO and EDF, another algorithm is developed called an adaptive algorithm which is best in both overloaded and under loaded condition. Then we compared EDF and GA based scheduling algorithm. EDF is used in under loaded condition and when the system becomes overloaded it changes to GA based algorithm. From the comparison of these two algorithms, we get another algorithm which is also called an adaptive algorithm. Performance of both algorithms is measured by using success rate, effective CPU utilization and execution time.

Computationally Efficient Method for the Generation of a Digital Terrain Model From Airborne LiDAR Data Using Connected Operators

This paper proposes a new mapping schema, named Θ mapping, for filtering nonground objects from LiDAR data, and the generation of a digital terrain model. By extending the CSL model, Θ mapping extracts the most contrasted connected-components from top-hat scale-space and attributes them for an adaptive multicriterion filter definition. Areas of the most contrasted connected-components and the standard deviations of contained points' levels are considered for this purpose. Computational efficiency is achieved by arranging the input LiDAR data into a grid, represented by a Max-Tree. Since a constant number of passes over the grid is required, the time complexity of the proposed method is linear according to the number of grid-cells. As confirmed by the experiments, the average CPU execution time decreases by nearly 98%, while the average accuracy improves by up to 10% in comparison with the related method.

An Empirical Study to Compare the Performance of some Symmetric and Asymmetric Ciphers

In this paper we present empirical results obtained from Java implementation of Elliptic curve Cryptosystem (ECC) as an asymmetric block cipher algorithm and a set of symmetric block cipher algorithms namely Triple-Data Encryption Standard (T-DES), Advanced Encryption Standard (AES), and Blowfish. Performance evaluation based on CPU execution time is presented under WinXP and Linux. We used in our implementation Java programming language, Java Cryptography Architecture (JCA) and Java Cryptography Extension (JCE). The evaluation of the performance of these algorithms is done for secret key generation and encryption and decryption operations. Results indicated that ECC outperforms the other encryption/decryption algorithms considered in this study regarding the security strength, speed, and key size of ECC. Also, ECC’s performance advantage increases as security needs increases for newly emerging applications.

Proportional 자원할당을 위한 OFDMA 시스템에서 채널 용량을 증대시키기 위한 향상된 전력 할당 기법

Orthogonal Frequency Division Multiple Access(OFDMA)는 4세대 무선 통신을 위해 고려되는 변조와 다중 접속 기술이다. 본 논문에서는, OFDMA 시스템에서 더 나은 proportional rate의 자원할당 기법을 위해 전체 전력, 비트 오류율, 유저 간의 rate proportionality에 대한 제한 조건을 만족시키면서, 유저의 전송률을 최대화 시키는 전략 할당 알고리즘에 대해 기술하였다. 그리고 subcarrier 할당의 비율과 유저의 정규화된 proportionality 조건에 기반을 둔 전력 할당 방식을 제안하였다. 유저에게 subcarrier 할당을 위한 방법으로는 greedy 알고리즘과 waterfilling 기술이 적용되었다. 제안된 알고리즘에 대한 평가는 시뮬레이션을 통해서 이루어졌으며, 시뮬레이션을 통해서 제안된 알고리즘이 유저 간의 rate proportionality를 유지하면서 더 높은 시스템 채널 용량과 적은 소요 시간을 보인다는 것을 확인할 수 있었다. The Orthogonal Frequency Division Multiple Access (OFDMA) is considered as a novel modulation and multiple access technique for 4th generation wireless systems. In this paper, we formulate a base station's power allocation algorithm for each user to maximize the user's sum rate, subject to constraints on total power, bit error rate, and rate proportionality among the users for a better proportional rate adaptive (RA) resource allocation method for OFDMA based system. We propose a novel power allocation method based on the proportion of subcarrier allocation and the user's normalized proportionality constant. We adapt a greedy algorithm and waterfilling technique for allocating the subcarriers among the users. In an end-to-end simulation, we validate that the proposed technique has higher system capacity and lower CPU execution times, while maintaining the acceptable rate proportionality among users.

Parallel option pricing on GPU: barrier options and realized variance options

This paper shows two examples of how the analysis of option pricing problems can lead to computational methods efficiently implemented in parallel. These computational methods outperform “general purpose” methods (i.e., for example, Monte Carlo, finite differences methods). The GPU implementation of two numerical algorithms to price two specific derivatives (continuous barrier options and realized variance options) is presented. These algorithms are implemented in CUDA subroutines ready to run on Graphics Processing Units (GPUs) and their performance is studied. The realization of these subroutines is motivated by the extensive use of the derivatives considered in the financial markets to hedge or to take risk and by the interest of financial institutions in the use of state of the art hardware and software to speed up the decision process. The performance of these algorithms is measured using the (CPU/GPU) speed up factor, that is using the ratio between the (wall clock) times required to execute the code on a CPU and on a GPU. The choice of the reference CPU and GPU used to evaluate the speed up factors presented is stated. The outstanding performance of the algorithms developed is due to the mathematical properties of the pricing formulae used and to the ad hoc software implementation. In the case of realized variance options when the computation is done in single precision the comparisons between CPU and GPU execution times gives speed up factors of the order of a few hundreds. For barrier options, the corresponding speed up factors are of about fifteen, twenty. The CUDA subroutines to price barrier options and realized variance options can be downloaded from the website http://www.econ.univpm.it/recchioni/finance/w13. A more general reference to the work in mathematical finance of some of the authors and of their coauthors is the website http://www.econ.univpm.it/recchioni/finance/.

Minimization of Delay for Query Processing in Peer to Peer Networks

Most of the research has been concentrated on sharing of files in Peer to Peer System .Our framework avoids the centralized structure of database management system and proposed the use of decentralized nature in a network. In the existing system, searching of neighbouring nodes is complex and provide more delay in communication of queries with no updation of mapping tables. This paper explains about the topological discovery of peers and creation of the acquaintances and mapping tables.Our objective is to implements a new environment in database by Query Processing system and explains how the execution and updation of query can be done in all the peers. In this paper, it deals with the effective query processing done with all the peers with respect to CPU execution time, creating an acquaintance, mapping tables with no communication delay in a large peer to peer system.

Parallelization and performance optimization on face detection algorithm with OpenCL: A case study

Face detect application has a real time need in nature. Although Viola-Jones algorithm can handle it elegantly, today's bigger and bigger high quality images and videos still bring in the new challenge of real time needs. It is a good idea to parallel the Viola-Jones algorithm with OpenCL to achieve high performance across both AMD and NVidia GPU platforms without bringing up new algorithms. This paper presents the bottleneck of this application and discusses how to optimize the face detection step by step from a very naive implementation. Some brilliant tricks and methods like CPU execution time hidden, stubbles usage of local memory as high speed scratchpad and manual cache, and variable granularity were used to improve the performance. Those technologies result in 4–13 times speedup varying with the image size. Furthermore, those ideas may throw on some light on the way to parallel applications efficiently with OpenCL. Taking face detection as an example, this paper also summarizes some universal advice on how to optimize OpenCL program, trying to help other applications do better on GPU.