摘要
针对大规模计算任务场景,提出在虚拟化环境下多GPU并行计算的实现方案,使用多线程或流处理的方式实现多GPU并行计算,并分析GPU多层次存储结构、传输等方面内容,采用经典的蒙特卡罗方法这一具有代表性的科学计算实例进行实验验证.
The solution of multi-GPU parallel computing under virtualization is proposed for the large-scale computing tasks. The multi-threads and flow mode are supported to implement the multi GPU cooperative computing. The multi-level storage structure for GPU, transmission and other aspects are analyzed to accelerate the progranx We make experimental verification of representative scientific computing examples such as Monte Carlo method. Experimental results show that with the increase of the calculation scale, it can achieve a speed-up ratio close to the number of GPU.
出处
《微电子学与计算机》
CSCD
北大核心
2016年第3期69-75,共7页
Microelectronics & Computer
基金
江苏省高校自然科学研究面上项目(14KJD510004)
关键词
GPU通用计算
虚拟化
并行计算
蒙特卡罗
general-purpose computation on GPU
virtualization
parallel computing
monte carlo
