Due to their highly parallel multi-cores architecture, GPUs are being increasingly used in a wide range of computationally intensive applications. Compared to CPUs, GPUs can achieve higher performances at accelerating...
详细信息
ISBN:
(纸本)9781479961238
Due to their highly parallel multi-cores architecture, GPUs are being increasingly used in a wide range of computationally intensive applications. Compared to CPUs, GPUs can achieve higher performances at accelerating the programs' execution in an energy-efficient way. Therefore GPGPU computing is useful for high performance computing applications and in many scientific research fields. In order to bring further performance improvements, GPU clusters are increasingly adopted. The energy consumed by GPUs cannot be neglected. Therefore, an energy-efficient time scheduling of the programs that are going to be executed by the parallel GPUs based on their deadline as well as the assigned priorities could be deployed to face their energetic avidity. For this reason, we present in this paper a model enabling the measure of the power consumption and the time execution of some elementary operations running on a single GPU using a new developed energy measurement protocol. Consequently, using our methodology, energy needs of a program could be predicted, allowing a better taskscheduling.
task scheduling algorithms in heterogeneous computing environment often provide limited performance owing to their low efficiency. This paper models and implements the first two phases of the Critical Path(CP) on a Pr...
详细信息
ISBN:
(纸本)9781510830363
task scheduling algorithms in heterogeneous computing environment often provide limited performance owing to their low efficiency. This paper models and implements the first two phases of the Critical Path(CP) on a Processor algorithm with the functional programming paradigm. Firstly, an overall research is conducted on the CPOP algorithms, then the logical relationship of the first two phrases of the algorithm is modeled based on Pi calculus theories, and finally the algorithm is implemented by n Pict programming language. The experimental results demonstrate that the implementation efficiency of the new programming algorithm is higher than that of the traditional C++ language. Therefore, Pi calculus can be applied to the three phases of the CPOP algorithm to improve the taskscheduling efficiency. This paper aims to prove that the CPOP algorithm can be more efficient than ever by using Pi calculus, through modeling and comparing the topological structures of the different types of taskscheduling models.
task scheduling algorithm in heterogeneous environment is programmed in a serial manner in the underlying by using traditional programming language. In this paper, we proposed that using the characteristics of paralle...
详细信息
task scheduling algorithm in heterogeneous environment is programmed in a serial manner in the underlying by using traditional programming language. In this paper, we proposed that using the characteristics of parallel computation of Pi-calculus, to parallel analysis and modeling of the heterogeneous earliest finish time(HEFT) algorithm based on pi-calculus. Then programmed it using PICT, which is a programming language that corresponds to the pi-calculus. By creating different topologies of the taskscheduling graph and compared with traditional programming language(C++ language) prove that the efficiency of using pi-calculus to develop the HEFT algorithm is higher than that of traditional programming language.
High throughput computing (HTC) is of great importance in grid computing environments. HTC is aimed at minimizing the total makespan of all of the tasks submitted to the grid environment in long execution of the syste...
详细信息
High throughput computing (HTC) is of great importance in grid computing environments. HTC is aimed at minimizing the total makespan of all of the tasks submitted to the grid environment in long execution of the system. To achieve HTC in grids, suitable task scheduling algorithms should be applied to dispatch the submitted tasks to the computational resources appropriately. In this paper, a new task scheduling algorithm is proposed to assign the tasks to the grid resources with goal of minimizing the total makespan of the environment. The proposed algorithm uses genetic approach to find the most suitable match between the tasks and grid resources. The simulation results obtained from applying the proposed algorithm to schedule independent and sequential tasks to the grid resources demonstrate the applicability of the algorithm in grid environments.
暂无评论