On the Use of Remote GPUs and Low-Power Processors for the Acceleration of Scientific Applications
On the Use of Remote GPUs and Low-Power Processors for the Acceleration of Scientific Applications
复制标题
关于使用远程 GPU 和低功耗处理器加速科学应用程序
DOI：
--
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
F. Silla
中科院分区：
文献类型：
--
作者：
Adrián Castelló;J. Duato;R. Mayo;Antonio J. Peña;E. Quintana;Vicente Roca;F. Silla
目前许多高性能集群在每个节点上都包含一个或多个GPU，以大幅减少应用程序的执行时间，但这些加速器的利用率通常远低于100%。在这种情况下，远程GPU虚拟化可以帮助降低采购成本以及整体能耗。在本文中，我们调查了几个“异构”的情况下，包括客户端GPU的节点运行CUDA应用程序和远程GPU配备的服务器节点提供访问NVIDIA硬件加速器的潜在开销和瓶颈。实验评估使用三个通用多核处理器（Intel Xeon，Intel Atom和ARM Cortex A9），两个图形加速器（NVIDIA GeForce GTX480和NVIDIA Quadro M1000）以及生物信息学和分子动力学模拟中产生的两个相关科学应用程序（CUDASW++和LAMMPS）进行。
Many current high-performance clusters include one or more GPUs per node in order to dramatically reduce application execution time, but the utilization of these accelerators is usually far below 100%. In this context, remote GPU virtualization can help to reduce acquisition costs as well as the overall energy consumption. In this paper, we investigate the potential overhead and bottlenecks of several “heterogeneous” scenarios consisting of client GPU-less nodes running CUDA applications and remote GPUequipped server nodes providing access to NVIDIA hardware accelerators. The experimental evaluation is performed using three general-purpose multicore processors (Intel Xeon, Intel Atom and ARM Cortex A9), two graphics accelerators (NVIDIA GeForce GTX480 and NVIDIA Quadro M1000), and two relevant scientific applications (CUDASW++ and LAMMPS) arising in bioinformatics and molecular dynamics simulations.
DOI：
10.1177/1094342010391989
发表时间：
2011-02-01
影响因子：
3.1
作者：
Dongarra, Jack;Beckman, Pete;Yelick, Kathy
通讯作者：
Yelick, Kathy