A Resource Selection System for Cycle Stealing in GPU Grids
A Resource Selection System for Cycle Stealing in GPU Grids
复制标题
DOI：
10.1007/s10723-008-9099-7
复制
发表时间：
2008-03
影响因子：
5.5
通讯作者：
Yuki Kotani;Fumihiko Ino;K. Hagihara
中科院分区：
文献类型：
--
作者：
Yuki Kotani;Fumihiko Ino;K. Hagihara
提出了一种在桌面网格环境中利用图形处理器作为通用计算资源的资源选择系统。我们的系统允许网格用户共享远程GPU，传统上，远程GPU专用于直接看到显示输出的本地用户。本文的主要贡献在于为非专用环境开发了这一新颖的系统。我们首先从网格用户的角度展示定义空闲GPU的标准。基于这些标准，我们的系统使用屏幕保护程序方法和一些传感器，以较低的开销检测空闲资源。这种较低开销的想法是为了避免在资源监控期间进行GPU干预。然后根据匹配服务选择检测到的空闲GPU，使系统适应GPU体系结构的快速发展。虽然系统本身还不能与当前的桌面网格系统互操作，但我们的想法可以应用到基于屏幕保护程序的系统，如BOINC。我们使用配备三代NVIDIA图形处理器的Windows PC对该系统进行了评估。实验结果表明，我们的系统实现了最高267ms的低开销，在最大化交付给网格用户的性能的同时，最小化了对本地用户的干扰。还在办公室环境中进行了一些案例研究，以证明该系统在检测到的空闲时间方面的有效性。
This paper presents a resource selection system for exploiting graphics processing units (GPUs) as general-purpose computational resources in desktop Grid environments. Our system allows Grid users to share remote GPUs, which are traditionally dedicated to local users who directly see the display output. The key contribution of the paper is to develop this novel system for non-dedicated environments. We first show criteria for defining idle GPUs from the Grid users’ point of view. Based on these criteria, our system uses a screensaver approach with some sensors that detect idle resources at a low overhead. The idea for this lower overhead is to avoid GPU intervention during resource monitoring. Detected idle GPUs are then selected according to a matchmaking service, making the system adaptive to the rapid advance of GPU architecture. Though the system itself is not yet interoperable with current desktop Grid systems, our idea can be applied to screensaver-based systems such as BOINC. We evaluate the system using Windows PCs with three generations of nVIDIA GPUs. The experimental results show that our system achieves a low overhead of at most 267 ms, minimizing interference to local users while maximizing the performance delivered to Grid users. Some case studies are also performed in an office environment to demonstrate the effectiveness of the system in terms of the amount of detected idle time.