Minimum Execution Time Research Articles

Power Optimized Task Scheduling using Genetic Algorithm (POTS-GA) in Cloud Environment

In a cloud environment, the allocation of tasks has become pivotal on account of rapid growth of user requests. The processing of user requests leads to a significant execution time, and a huge amount of power is also consumed. Consequently, task scheduling for optimizing makes pan and power usage has become critical, particularly in a heterogeneous environment. This research work proposes Power-Optimized Task Scheduling using Genetic Algorithm (POTS-GA) that aims to minimize execution time and power consumption. The proposed strategy employs genetic algorithm to take scheduling decision while taking into consideration the execution time and overall power consumption of resources. The fitness computation considering both objectives and the customized genetic operators ensure to search for a better scheduling solution. The experiments performed on a large number of tasks and virtual machines show that the proposed POTS-GA approach outperforms other task scheduling strategies including Efficient Task Allocation using Genetic Algorithm (ETA-GA), Round Robin algorithm (RRA), and First Come First Serve (FCFS) and Greedy algorithm in terms of makes pan and power consumption.

Predictive Analysis for Personalized Machine: Leveraging Patient Data for Enhanced Healthcare

This research explores predictive analysis for personalized machine: leveraging patient data for enhanced healthcare. By leveraging the power of information and analytics, the healthcare industry can be driven towards a more patient-centric, proactive model that enhances outcomes and improve the overall quality of care. The objectives of the study are to: determine the significance and challenges of predictive analytics in healthcare, ascertain the data analytics techniques used in healthcare to enhance patient care, find out how predictive analytics can be applied for enhanced healthcare, and determine the ethical considerations associated with healthcare predictive analytics. This study employs the case study approach and experimental design. The study analyzes case studies of real-time deployment of predictive analytics models in healthcare centers, examines how these models enhance the healthcare delivery in those centers. Experiments were also conducted to understand how predictive analytics works. The C4.5 learning algorithm was employed to predict the presence of chronic kidney disease (CKD) in patients and differentiate between those not affected by the condition. The C4.5 classifier shows reasonable strength, evident in the large number of rightly classified occurrences (396) and a low misclassification of only 4 occurrences. This is further demonstrated by a low error rate of 0.37, as shown in table 5. The prevalence of this algorithm is emphasized by the large value of KS (0.97), indicating the classifier’s ground-breaking accuracy and performance. The performance of C4.5, featured by its minimal execution time and accuracy, puts it as a decent classifier. This characteristic makes it specifically well-suited for application in the healthcare sector, particularly for tasks involving prediction and classification. The application of data analytics methods for predictive analysis holds significant benefits in the health sector, as it gives us the power to predict and address potential threats to human health, covering different age groups, from the young ones to the elderly. This proactive method enables early disease detection, helping in timely interventions and contributing to better decision-making.

Test case prioritization based on fault sensitivity analysis using ranked NSGA-2

This paper discusses regression testing in software maintenance, focusing on test case prioritization to verify modifications to software functionality efficiently. The primary goal is to rank test cases, prioritizing those covering more code or faults with minimal execution time. The challenge lies in prioritizing numerous test cases generated during development and maintenance. Various algorithms, including greedy approaches and meta-heuristic techniques, address this challenge. The paper introduces a ranking-based non-dominated sorting genetic algorithm (NSGA-2) for test case prioritization, emphasizing cases sensitive to faults caused by modifications or new functionality. Historical data is prioritized, with key objectives including the sensitive index, execution cost, and average percentage of fault detection (APFD). The proposed model is tested on handcrafted and benchmark Java-based applications, comparing its performance to state-of-the-art algorithms in test case prioritization.

On Batch Writing in COTS RFID Systems

We study the batch writing problem in RFID systems, where the reader seeks the most time-efficient way to write information into a given subset of tags. The problem is analogous to the multicast problem in classical networks, but one-to-many transmission is not supported in COTS RFID systems. Driven by the technical challenge, this paper addresses the problem of designing batch writing algorithms for COTS RFID systems. We make three contributions. Firstly, we establish the minimal execution time for any batch writing algorithm, thus setting the theoretical performance limit. Secondly, we quantitatively compare and gauge the existing propositions applicable to our problem. Thirdly, we develop a novel batch writing algorithm with minimal 25% performance gain over the best state-of-the-art solution. Our key technicalities are designing an encoding scheme allowing the reader to efficiently perform batch writing and optimizing the batch writing sequence to minimize the overall execution time. We also perform extensive experiments to demonstrate the effectiveness of our algorithm.

An asymmetric S-curve trajectory planning based on an improved jerk profile

AbstractIn this paper, a method of planning the expanded S-curve trajectory of robotic manipulators is proposed to minimize the execution time as well as to achieve the smoother trajectory generation in the deceleration stage for point-to-point motions. An asymmetric parameter is added to the piecewise sigmoid function for an improved jerk profile. This asymmetric profile is continuous and infinitely differentiable. Based on this profile, two analytical algorithms are presented. One is applied to determine the suitable time intervals of trajectory satisfying the time optimality under the kinematic constraints, and the other is to determine the asymmetric parameter generating the minimum execution time. Also, the calculation procedure for the time-scaled synchronization for all joints is given to decrease unnecessary loads onto the actuators. The velocity, acceleration, jerk and snap (the derivative of jerk) of the joints and the end-effector are equal to zero at two end points of motion. The simulation results through 3 DOF and 6 DOF robotic manipulators show that our approach reduces the jerk and snap of the deceleration stage effectively while decreasing the total execution time. Also, the analysis for a single DOF mass-spring-damper system indicates that the residual vibration could be reduced to 10% more than the benchmark techniques in case velocity, acceleration and jerk are limited to 1.24 m/s, 6 m/s2 and 80 m/s3, respectively and displacement is set to 0.8m. These results manifest that the performance of reducing residual vibrations is good and demonstrate an important characteristic of the proposed profile suitable for point-to-point motion.

Tsoa: a two-stage optimization approach for GCC compilation options to minimize execution time

Tsoa: a two-stage optimization approach for GCC compilation options to minimize execution time

Optimizing Task Scheduling in Cloud Data Centres with Dynamic Resource Allocation Using Genetic Algorithm (TSOGA)

Nowadays, Massive business applications are increasingly giving attention to cloud computing data centres because of its high potential, adaptability, and efficiency in supplying several sources of both software and hardware to support networked consumers. The criteria for autonomy of virtual machines necessitate a flexible resource allocation strategy for Virtual Machines (VMs) .The majority of resource utilization models were inaccurate, making it impossible to determine the virtual machine's energy usage directly from the hardware. Due to the size of modern data centres and the constantly changing character of their resource supply, efficient scheduling solutions must be developed to oversee these resources and meet the objectives of both cloud service providers and cloud customers. Hence an algorithm called Task Scheduling Optimization based Genetic Algorithm (TSOGA) has been proposed to dynamically allocate the resources in pursuit of scheduling the tasks in cloud data centers. The proposed module initially focuses on task scheduling process, followed by optimized running time of task execution. For data centres with dynamic resource allocation, the goal of TSOGA is to efficiently assign jobs to resources while minimizing execution time and optimizing resource utilization. Thus, to manage the data centres while achieving high levels of efficiency in resource allocation, we constructed a virtual node for our research. Incorporation of Genetic Algorithm is to determine an ideal or nearly ideal schedule for carrying out tasks using the available resources while taking into account a variety of restrictions and goals, such as minimizing execution and waiting time of task during dynamic scheduling process and efficient resource utilization.

Parallelization with load balancing of the weather scheme WSM7 for heterogeneous CPU-GPU platforms

This article provides an enhanced parallelization of the WSM7 microphysics scheme for the Weather Research and Forecasting Model (WRF). The parallelization is designed to maximize the utilization of a heterogeneous computing system consisting of CPUs, GPUs or both. Therefore the reference implementation of the WSM7 scheme is re-implemented for the heterogeneous execution model. For each time step, a dynamic load distribution is introduced which balances the computational load between the two components aiming for an overall minimum execution time. The evaluation of the parallelized implementation is done for a specific weather situation. Specifically, the precipitation of the low-pressure zone “Bernd” from July 2021 is simulated using an Intel Core i7-7700 CPU and a NVIDIA GTX 1070 GPU. The results show a speedup of up to 28.51 for the GPU version in comparison with the reference implementation. The heterogeneous dynamic load balancing increases the speedup achieved even further by introducing a distribution factor that is updated for each time step.

Optimal Model Partitioning with Low-Overhead Profiling on the PIM-based Platform for Deep Learning Inference

Recently Processing-in-Memory (PIM) has become a promising solution to achieve energy-efficient computation in data-intensive applications by placing computation near or inside the memory. In most Deep Learning (DL) frameworks, a user manually partitions a model’s computational graph (CG) onto the computing devices by considering the devices’ capability and the data transfer. The Deep Neural Network (DNN) models become increasingly complex for improving accuracy; thus, it is exceptionally challenging to partition the execution to achieve the best performance, especially on a PIM-based platform requiring frequent offloading of large amounts of data. This article proposes two novel algorithms for DL inference to resolve the challenge: low-overhead profiling and optimal model partitioning. First, we reconstruct CG by considering the devices’ capability to represent all the possible scheduling paths. Second, we develop a profiling algorithm to find the required minimum profiling paths to measure all the node and edge costs of the reconstructed CG. Finally, we devise the model partitioning algorithm to get the optimal minimum execution time using the dynamic programming technique with the profiled data. We evaluated our work by executing the BERT, RoBERTa, and GPT-2 models on the ARM multicores with the PIM-modeled FPGA platform with various sequence lengths. For three computing devices in the platform, i.e., CPU serial/parallel and PIM executions, we could find all the costs only in four profile runs, three for node costs and one for edge costs. Also, our model partitioning algorithm achieved the highest performance in all the experiments over the execution with manually assigned device priority and the state-of-the-art greedy approach.

О времени реализации распределенных вычислений в синхронном режиме при ограниченном числе копий программного ресурса

The article constructs a mathematical model of distributed computing with a limited number of copies of a structured software resource; in cases of unlimited and limited parallelism by the number of processors of a multiprocessor distributed computing system, the problems of finding the minimum execution time of heterogeneous, homogeneous and identically distributed competing processes in a synchronous mode, ensuring the continuous execution of each block of software resource by all processes.

Machine Learning Algorithms for Intrusion Detection in IoT Prediction and Performance Analysis

The rapid growth of the Internet of Things (IoT) has created a large attack surface for adversaries to launch destructive cyber attacks. Intrusion Detection Systems (IDS) can be used to identify and monitor anomalies and threats in the IoT system. In this paper, we compare the performance of various machine learning algorithms for building IDS, including decision tree (DT), random forest (RF), k-Nearest Neighbor (k-NN), Ada Boost, and support vector machine (SVM), using the IoTID20 dataset. This dataset has three target classes, including a binary class for normal or abnormal behavior and classes for categories and sub-categories of the binary class. We select the most relevant features to minimize execution time and improve accuracy. Our results show that the decision tree algorithm generally achieved the highest accuracy at 99.80% with the lowest error rate compared to the other algorithms.

Internet of Things Data Cloud Jobs Scheduling Using Modified Distance Cat Swarm Optimization

IoT cloud computing provides all functions of traditional computing as services through the Internet for the users. Big data processing is one of the most crucial advantages of IoT cloud computing. However, IoT cloud job scheduling is considered an NP-hard problem due to the hardness of allocating the clients’ jobs to suitable IoT cloud provider resources. Previous work on job scheduling tried to minimize the execution time of the job scheduling in the IoT cloud, but it still needs improvement. This paper proposes an enhanced job scheduling mechanism using cat swarm optimization (CSO) with modified distance to minimize the execution time. The proposed job scheduling mechanism first creates a set of jobs and resources to generate the population by randomly assigning the jobs to resources. Then, it evaluates the population using the fitness value, which represents the execution time of the jobs. In addition, we use iterations to regenerate populations based on the cat’s behaviour to produce the best job schedule that gives the minimum execution time for the jobs. We evaluated the proposed mechanism by implementing an initial simulation using Java Language and then conducted a complete simulation using the CloudSim simulator. We ran several experimentation scenarios using different numbers of jobs and resources to evaluate the proposed mechanism regarding the execution time. The proposed mechanism significantly reduces the execution time when we compare the proposed mechanism against the firefly algorithm and glowworm swarm optimization. The average execution time of the proposed cat swarm optimization was 131, while the average execution times for the firefly algorithm and glowworm optimization were 237 and 220, respectively. Hence, the experimental findings demonstrated that the proposed mechanism performs better than the firefly algorithm and glowworm swarm optimization in reducing the execution time of the jobs.

Intelligent Cloudlet Scheduling for Optimized Execution Time in Cloud Computing Environments

Cloud computing has become a cornerstone of modern IT infrastructure, offering scalable and flexible resources. However, efficient resource management, particularly cloudlet scheduling, presents a significant challenge due to its NP-hard nature. This paper introduces a novel heuristic-based cloudlet scheduling algorithm aimed at minimizing execution time and improving load balancing in cloud computing environments. We detail the development and implementation of the algorithm, along with a simulation setup using the CloudSim toolkit to evaluate its performance against existing methods. Results from extensive simulations demonstrate that the proposed algorithm consistently reduces turnaround times, thus optimizing resource allocation. The findings suggest that our approach can significantly impact cloud computing efficiency, paving the way for improved service provider offerings and user satisfaction. The implications of these advancements are discussed, alongside potential directions for future research in dynamic cloud environments.

RACE Resource Aware Cost-Efficient Scheduler for Cloud Fog Environment

Abstract: Fog computing is one of the new computing structures which takes the Cloud to the verge of the network. The structure is formulated for applications that need low latency. Fog Computing has been projected to improve the disadvantages of Cloud Computing. The system is confronted with the variability of dynamic resources that are heterogeneous and distributed. Hence, efficient scheduling and resource allocation are necessary to maximize the use of these resources and the satisfaction of users. In this paper, a resource-aware scheduler RACE (Resource Aware Cost-Efficient Scheduler) is proposed to distribute the incoming application modules to Fog devices that maximize resource utilization at the Fog layer, reduces the monetary cost of using Cloud resources with minimum execution time of applications and minimum bandwidth usage. This RACE comprises of two algorithms. The Module Scheduler in RACE categorizes the incoming application modules according to their computation and bandwidth requirements which are then placed by Compare Module. Comprehensive experimental results obtained from the simulation by using ifogsim simulator show that our approach performs better in most of the cases as compared to the Traditional Cloud placement and the baseline algorithm.

An efficient intelligent task management in autonomous vehicles using AIIOT and optimal kernel adaptive SVM

The fast evolution of artificial intelligence (AI) and the IoT gained more interest in the development of autonomous vehicles. The main challenges faced by autonomous car manufacturers are high computation costs and the lag of intelligent task management systems. The accident rates created by autonomous vehicles are increasing rapidly due to their unrestrained traffic, inaccurate location, and mapping methods. So, secure driving becomes the main concern in self-driving vehicle design. Moreover, the inadequate battery life and computation power made the system complex to minimize execution time as well as resource computation. Therefore, to handle all these complications, an intelligent task-managing system for autonomous vehicles is proposed in this paper. In this, each task is optimally executed by invoking the supervised resource predictor kernel data adaptive support vector machine-based multimodal bacterial foraging (KDASVM-MBF) method. The KDASVM-MBF intelligent task scheduling method is proposed to distribute all the tasks to the suitable processor based on central processing unit (CPU) usage and emergency. The proposed model is implemented in Python IDE-version 3.8 and examined using two multicore processors (Nvidia and AIIoT). The potential capability of the introduced type is evaluated by computing the performance methods such as response time, resource utilization, CPU utilization, execution time, prediction accuracy, and task miss rate. The experimental results reveal that the established KDASVM-MBF method accomplishes prediction accuracy of about 97% and 98% for Nvidia and AIIoT processors respectively with minimum task miss rate and execution time.

A multithreaded CUDA and OpenMP based power‐aware programming framework for multi‐node GPU systems

SummaryIn the article, we have proposed a framework that allows programming a parallel application for a multi‐node system, with one or more graphical processing units (GPUs) per node, using an OpenMP+extended CUDA API. OpenMP is used for launching threads responsible for management of particular GPUs and extended CUDA calls allow to transfer data and launch kernels on local and remote GPUs. The framework hides inter‐node MPI communication from the programmer. For optimization, the implementation takes advantage of the MPI_THREAD_MULTIPLE mode allowing: multiple threads handling distinct GPUs as well as overlapping communication and computations transparently using multiple CUDA streams. The solution allows data parallelization across available GPUs in order to minimize execution time and supports a power‐aware mode in which GPUs are automatically selected for computations using a greedy approach in order not to exceed an imposed power limit. We have implemented and benchmarked three parallel applications including: finding the largest divisors; verification of the Collatz conjecture; finding patterns in vectors. These were tested on three various systems: a GPU cluster with 16 nodes, each with NVIDIA GTX 1060 GPU; a powerful 2‐node system—one node with 8 NVIDIA Quadro RTX 6000 GPUs, the second with 4 NVIDIA Quadro RTX 5000 GPUs; a heterogeneous environment with one node with 2 NVIDIA RTX 2080 and 2 nodes with NVIDIA GTX 1060 GPUs. We demonstrated effectiveness of the framework through execution times versus power caps within ranges of 100–1400 W, 250–3000 W, and 125–600 W for these systems respectively as well as gains from using two versus one CUDA streams per GPU. Finally, we have shown that for the testbed applications the solution allows to obtain high speed‐ups between 89.3% and 97.4% of the theoretically assessed ideal ones, for 16 nodes and 2 CUDA streams, demonstrating very good parallel efficiency.

Deep Learning-based Resource Prediction and Mutated Leader Algorithm Enabled Load Balancing in Fog Computing

Load balancing plays a major part in improving the performance of fog computing, which has become a requirement in fog layer for distributing all workload in equal manner amongst the current Virtual machines (VMs) in a segment. The distribution of load is a complicated process as it consists of numerous users in fog computing environment. Hence, an effectual technique called Mutated Leader Algorithm (MLA) is proposed for balancing load in fogging environment. Firstly, fog computing is initialized with fog layer, cloud layer and end user layer. Then, task is submitted from end user under fog layer with cluster of nodes. Afterwards, load balancing process is done in each cluster and the resources for each VM are predicted using Deep Residual Network (DRN). The load balancing is accomplished by allocating and reallocating the task from the users to the VMs in the cloud based on the resource constraints optimally using MLA. Here, the load balancing is needed for optimizing resources and objectives. Lastly, if VMs are overloaded and then the jobs are pulled from associated VM and allocated to under loaded VM. Thus the proposed MLA achieved minimum execution time is 1.472ns, cost is $69.448 and load is 0.0003% respectively.

SAGAS: Simulated annealing and greedy algorithm scheduler for laboratory automation.

During laboratory automation of life science experiments, coordinating specialized instruments and human experimenters for various experimental procedures is important to minimize the execution time. In particular, the scheduling of life science experiments requires the consideration of time constraints by mutual boundaries (TCMB) and can be formulated as the "scheduling for laboratory automation in biology" (S-LAB) problem. However, existing scheduling methods for the S-LAB problems have difficulties in obtaining a feasible solution for large-size scheduling problems at a time sufficient for real-time use. In this study, we proposed a fast schedule-finding method for S-LAB problems, SAGAS (Simulated annealing and greedy algorithm scheduler). SAGAS combines simulated annealing and the greedy algorithm to find a scheduling solution with the shortest possible execution time. We have performed scheduling on real experimental protocols and shown that SAGAS can search for feasible or optimal solutions in practicable computation time for various S-LAB problems. Furthermore, the reduced computation time by SAGAS enables us to systematically search for laboratory automation with minimum execution time by simulating scheduling for various laboratory configurations. This study provides a convenient scheduling method for life science automation laboratories and presents a new possibility for designing laboratory configurations.

NLBBODE optimizer for accurate and fast modeling of photovoltaic module/string generator and its application to solve real-world constrained optimization problems

In this paper, a new optimizer is presented to quickly and accurately identify parameters of the photovoltaic (PV) module/string models. This optimizer is named Nested Loop Biogeography-based Optimization - Differential Evolution referred to as (NLBBODE). It has been developed to identify the PV parameters with reasonable computational effort and minimum execution time, despite the nonlinearity of the PV system dynamics and the insufficiency of data. In addition, the NLBBODE optimizer is used to solve some engineering design problems known as highly constrained, nonlinear, and non-convex. The weaknesses of the original versions of BBO and DE approaches have been overpassed. Furthermore, the proposed optimizer is compared to the state-of-the-art metaheuristic methods using performance evaluation metrics. The computational resources needed to obtain the optimum solution using NLBBODE are significantly reduced due to the nested loop design. The results obtained prove that the NLBBODE optimizer is a suitable candidate to solve the problem of the PV modeling as well as to solve various real-world constrained optimization problems, with high accuracy and low processor runtime, which is a necessary condition for online applications. For instance, the well-known Photowatt-PWP-201 module model represented in the single diode model, NLBBODE registers a standard deviation value of 1.4682E-17 within a time of 13.9 s. For the STM6-40/36 module model represented in the single diode model, the NLBBODE optimizer records a standard deviation value of 6.191583E-18 within an execution time of 6 s. For the speed reducer design problem, the standard deviation obtained by the NLBBODE is 1.515824E-13 and needs a time of less than 1 s to obtain the optimum solution.

A hybrid learning frame work for recognition abnormal events intended from surveillance videos

The need for a monitoring system has grown as a result of rising crime and anomalous activity. To avoid unusual incidents, the common man initiated video surveillance of important areas, which was then passed on to the government. In typical surveillance operations, surveillance devices create a vast volume of data that must be manually analysed. Manually handling huge data sets in real time results in information loss. To prevent abnormal incidents, the actions in sensitive areas can be properly monitored, evaluated, and alerted to the appropriate authorities. Previous deep learning-based activity identification methods have appeared, but the findings are inaccurate, and the proposed Hybrid Machine Learning Algorithms (HMLA) incorporate two detection methods for surveillance videos like as Transfer Learning (TL) and Continual Learning (CL). As a result, the suspicious activity in the video may be missed. Consequently, numerous image processing and computer vision technologies were used in activity detection to decrease human effort and mistakes in surveillance operations. Activities in sensitive areas can be properly monitored and evaluated to avoid unusual incidents, and the appropriate authorities may be alerted. Hence, in order to decrease human error and effort in surveillance operations, activity recognition embraced a variety of image processing and computer vision technologies. In this present work, the capacity has constraints that impact recognition accuracy. Consequently, this research paper presents a HMLA based technique that uses feature extraction using multilayer (Long Short Term Memory) LSTM, Convolutional Neural Networks (CNN), and Temporal feature extraction using multilayer LSTM to improve identification accuracy by 96% while requiring minimal execution time. To show the superior performance of the proposed hybrid machine learning technique, a standard UCF crime dataset was utilised for experimental analysis and compared to existing deep learning algorithms.