List of Articles Scheduling

• Open Access Article
  • Abstract Page
  • Full-Text
  
  1 - Scheduling tasks in cloud environments using mapping framework - reduction and genetic algorithm
  nima khezr nima jafari novimipour
  20.1001.1.27170414.1397.10.37.2.7
  Task scheduling is a vital component of any distributed system such as grids, clouds, and peer-to-peer networks that refer tasks to appropriate resources for execution. Common scheduling methods have disadvantages such as high time complexity, inconsistent execution of More

  Task scheduling is a vital component of any distributed system such as grids, clouds, and peer-to-peer networks that refer tasks to appropriate resources for execution. Common scheduling methods have disadvantages such as high time complexity, inconsistent execution of input tasks, and increased program execution time. Exploration-based scheduling algorithms to prioritize tasks from Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  2 - A Task Mapping and Scheduling Algorithm based on Genetic Algorithm for Embedded System Design
  mohadese nikseresht Mohsen Raji
  20.1001.1.27170414.1400.13.47.12.8
  Embedded system designers face numerous design requirements and objectives (such as runtime, power consumption and reliability). Since meeting one of these requirements mostly contradicts other design requirements, it seem to be inevitable to apply multi-objective appr More

  Embedded system designers face numerous design requirements and objectives (such as runtime, power consumption and reliability). Since meeting one of these requirements mostly contradicts other design requirements, it seem to be inevitable to apply multi-objective approaches in various stages of designing embedded systems, including task scheduling step. In this paper, a multi-objective task mapping and scheduling in the design stage of the embedded system is presented. In this method, tasks are represented by task graphs assuming that the hardware architecture platform is given and determined. In order to manage the dependencies between tasks in the task graph, a segmentation method is used, in which the tasks that can be executed simultaneously are specified in a segment and is considered in the scheduling process. In the proposed method, the task mapping and scheduling problem is modeled as a genetic algorithm-based multi-objective optimization problem considering execution time, energy consumption, and reliability. In comparison to similar previous works, the proposed scheduling method respectively provides 21.4%, 19.2%, and 20% improvement in execution time, energy consumption, and reliability. Applying a multi-objective helps the designer to pick out the best outcome according to different considerations. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  3 - A Multi-Objective Differential Evolutionary Algorithm-based Approach for Resource Allocation in Cloud Computing Environment
  Saeed Bakhtiari Mahan Khosroshahi
  20.1001.1.27170411.1401.14.51.9.5
  In recent years, the cloud computing model has received a lot of attention due to its high scalability, reliability, information sharing and low cost compared to separate machines. In the cloud environment, scheduling and optimal allocation of tasks affects the effectiv More

  In recent years, the cloud computing model has received a lot of attention due to its high scalability, reliability, information sharing and low cost compared to separate machines. In the cloud environment, scheduling and optimal allocation of tasks affects the effective use of system resources. Currently, common methods for scheduling in the cloud computing environment are performed using traditional methods such as Min-Min and meta-heuristic methods such as ant colony optimization algorithm (ACO). The above methods focused on optimizing one goal and do not estimate multiple goals at the same time. The main purpose of this research is to consider several objectives (total execution time, service level agreement and energy consumption) in cloud data centers with scheduling and optimal allocation of tasks. In this research, multi-objective differential evolution algorithm (DEA) is used due to its simple structure features and less adjustable parameters. In the proposed method, a new approach based on DEA to solve the problem of allocation in cloud space is presented which we try to be effective in improving resource efficiency and considering goals such as time, migration and energy by defining a multi-objective function and considering mutation and crossover vectors. The proposed method has been evaluated through a CloudSim simulator by testing the workload of more than a thousand virtual machines on Planet Lab. The results of simulation show that the proposed method in comparison with IqrMc, LrMmt and FA algorithms, in energy consumption by an average of 23%, number of migrations by an average of 29%, total execution time by an average of 29% and service level agreement violation (SLAV) by an average of 1% has been improved. In this case, use of the proposed approach in cloud centers will lead to better and appropriate services to customers of these centers in various fields such as education, engineering, manufacturing, services, etc. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  4 - Fuzzy Multicore Clustering of Big Data in the Hadoop Map Reduce Framework
  Seyed Omid Azarkasb Seyed Hossein Khasteh Mostafa  Amiri
  20.1001.1.27170411.1402.15.55.5.6
  A logical solution to consider the overlap of clusters is assigning a set of membership degrees to each data point. Fuzzy clustering, due to its reduced partitions and decreased search space, generally incurs lower computational overhead and easily handles ambiguous, no More

  A logical solution to consider the overlap of clusters is assigning a set of membership degrees to each data point. Fuzzy clustering, due to its reduced partitions and decreased search space, generally incurs lower computational overhead and easily handles ambiguous, noisy, and outlier data. Thus, fuzzy clustering is considered an advanced clustering method. However, fuzzy clustering methods often struggle with non-linear data relationships. This paper proposes a method based on feasible ideas that utilizes multicore learning within the Hadoop map reduce framework to identify inseparable linear clusters in complex big data structures. The multicore learning model is capable of capturing complex relationships among data, while Hadoop enables us to interact with a logical cluster of processing and data storage nodes instead of interacting with individual operating systems and processors. In summary, the paper presents the modeling of non-linear data relationships using multicore learning, determination of appropriate values for fuzzy parameterization and feasibility, and the provision of an algorithm within the Hadoop map reduce model. The experiments were conducted on one of the commonly used datasets from the UCI Machine Learning Repository, as well as on the implemented CloudSim dataset simulator, and satisfactory results were obtained.According to published studies, the UCI Machine Learning Repository is suitable for regression and clustering purposes in analyzing large-scale datasets, while the CloudSim dataset is specifically designed for simulating cloud computing scenarios, calculating time delays, and task scheduling. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  5 - WSTMOS: A Method For Optimizing Throughput, Energy, And Latency In Cloud Workflow Scheduling
  Arash Ghorbannia Delavar Reza Akraminejad sahar mozafari
  20.1001.1.27170411.1402.15.57.5.0
  Application of cloud computing in different datacenters around the world has led to generation of more co2 gas. In addition, energy and throughput are the two most important issues in this field. This paper has presented an energy and throughput-aware algorithm for sche More

  Application of cloud computing in different datacenters around the world has led to generation of more co2 gas. In addition, energy and throughput are the two most important issues in this field. This paper has presented an energy and throughput-aware algorithm for scheduling of compressed-instance workflows in things-internet by cluster processing in cloud. A method is presented for scheduling cloud workflows with aim of optimizing energy, throughput, and latency. In the proposed method, time and energy consumption has been improved in comparison to previous methods by creating distance parameters, clustering inputs, and considering real execution time. In WSTMOS method by considering special parameters and real execution time, we managed to reach the optimized objective function. Moreover, in the proposed method parameter of time distance of tasks to virtual machines for reduction of number of migration in virtual machines was applied. In WSTMOS method by organizing the workflow inputs to low, medium and heavy groups and also by distributing appropriate load on more suitable servers for processors threshold, we accomplished to optimize energy and cost. Energy consumption was reduced by 4.8 percent while the cost was cut down by 4.4 percent using this method in comparison to studied method. Finally, average delay time, power and workload are optimized in comparison to previous methods. Manuscript profile
• Open Access Article
  • Abstract Page
  • Full-Text
  
  6 - Improving the load balancing in Cloud computing using a rapid SFL algorithm (R-SFLA)
  Kiomars Salimi Mahdi Mollamotalebi
  20.1001.1.27170411.1402.15.57.13.8
  Nowadays, Cloud computing has many applications due to various services. On the other hand, due to rapid growth, resource constraints and final costs, Cloud computing faces with several challenges such as load balancing. The purpose of load balancing is management of th More

  Nowadays, Cloud computing has many applications due to various services. On the other hand, due to rapid growth, resource constraints and final costs, Cloud computing faces with several challenges such as load balancing. The purpose of load balancing is management of the load distribution among the processing nodes in order to have the best usage of resources while having minimum response time for the users’ requests. Several methods for load balancing in Cloud computing have been proposed in the literature. The shuffled frog leaping algorithm for load balancing is a dynamic, evolutionary, and inspired by nature. This paper proposed a modified rapid shuffled frog leaping algorithm (R-SFLA) that converge the defective evolution of frogs rapidly. In order to evaluate the performance of R-SFLA, it is compared to Shuffled Frog Leaping Algorithm (SFLA) and Augmented Shuffled Frog Leaping Algorithm (ASFLA) by the overall execution cost, Makespan, response time, and degree of imbalance. The simulation is performed in CloudSim, and the results obtained from the experiments indicated that the proposed algorithm acts more efficient compared to other methods based on the above mentioned factors. Manuscript profile