arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1907.09272cs.DChep-ex
收藏

扩展ARC信息提供方以报告有关GPU资源的信息

Extending the ARC Information Providers to report information on GPU resources

Max Isacson, Mattias Ellert, Richard Brenner

摘要:

图形处理单元上的通用计算(GPGPU)已被引入许多科学研究领域,例如生物信息学、密码学、计算机视觉和深度学习。然而,高能物理(HEP)社区的计算模型仍然主要围绕传统的CPU资源。轨迹拟合、粒子重建和蒙特卡罗模拟等任务可以从高吞吐量GPGPU计算模型中受益匪浅,从而简化分析周转中的瓶颈。本技术说明描述了在GRID中间件中实现集成GPU发现机制以促进GPGPU的基础。

英文摘要:

General-purpose Computing on Graphics Processing Units (GPGPU) has been introduced to many areas of scientific research such as bioinformatics, cryptography, computer vision, and deep learning. However, computing models in the High-energy Physics (HEP) community are still mainly centred around traditional CPU resources. Tasks such as track fitting, particle reconstruction, and Monte Carlo simulation could benefit greatly from a high-throughput GPGPU computing model, streamlining bottlenecks in analysis turnover. This technical note describes the basis of an implementation of an integrated GPU discovery mechanism in GRID middleware to facilitate GPGPU.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑