| CPU-GPU异构多核系统的动态任务调度算法 |
| |
| 引用本文： | 裴颂文,宁静,张俊格. CPU-GPU异构多核系统的动态任务调度算法[J]. 计算机应用研究, 2016, 33(11) |
| |
| 作者姓名： | 裴颂文 宁静 张俊格 |
| |
| 作者单位： | 上海理工大学 光电信息与计算机工程学院,上海理工大学 光电信息与计算机工程学院,上海理工大学 光电信息与计算机工程学院 |
| |
| 基金项目： | 上海市自然科学基金(15ZR1428600),计算机体系结构国家重点实验室开放资助项目(CARCH201206) |
| |
| 摘 要： | CPU-GPU异构多核系统对计算密集型的应用加速效果显著而得到广泛应用,但该系统易出现负载均衡问题。针对此问题,本文提出了一种CPU-GPU异构多核系统的动态任务调度算法。该算法充分利用CPU的线程资源和GPU的计算资源,准确测量CPU和GPU的计算能力,从而动态调整分配到CPU和GPU上的数据块大小,减小负载的总执行时间,提高系统加速比。实验结果表明,该算法使得系统加速比提高34%~103%。
|
| 关 键 词： | 动态调度 负载均衡 自适应分配 异构计算 |
| 收稿时间： | 2015-07-20 |
| 修稿时间： | 2016-09-13 |
A dynamic task scheduling algorithm based on CPU-GPU heterogeneous multi-core system |
| |
| PEI Song-wen,NING Jing and ZHANG Jun-ge. A dynamic task scheduling algorithm based on CPU-GPU heterogeneous multi-core system[J]. Application Research of Computers, 2016, 33(11) |
| |
| Authors: | PEI Song-wen NING Jing ZHANG Jun-ge |
| |
| Affiliation: | Department of Computer Science and Engineering,University of Shanghai for Science and Technology,Department of Computer Science and Engineering,University of Shanghai for Science and Technology,Department of Computer Science and Engineering,University of Shanghai for Science and Technology |
| |
| Abstract: | CPU-GPU heterogeneous multi-core system has been widely applied because of its acceleration effects for compute-intensive applications. However, the problem of workload imbalance is serious. Therefore, we proposed a Dynamic Task Scheduling Algorithm (DTSA) based on CPU-GPU heterogeneous multi-core system.In order to guarantee that all cores are doing useful work it makes full use of CPU and GPU. Furthermore, it can accurately measure the computational power of GPUs and CPUs respectively, dynamically adjust the size of data blocks to be executed on CPUs and GPUs, and finally reduce the total executing time of workloads and increases the system speedup. According to the results of experiments by using this algorithm, the system speedup increases by 34%~103%. |
| |
| Keywords: | dynamic scheduling workload balance adaptive allocation heterogeneous computing |
|
| 点击此处可从《计算机应用研究》浏览原始摘要信息 |
|
点击此处可从《计算机应用研究》下载全文 |
|
