arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.03963cs.DC
收藏

将IceCube GPU计算扩展到云

Expanding IceCube GPU computing into the Clouds

Igor Sfiligoi, Shava Smallen, Frank Würthwein, Nicole Wolter, David Schultz, Benedikt Riedel

摘要:

IceCube协作依赖GPU计算来满足其许多需求,包括光线追踪模拟和机器学习活动。然而,GPU在科学资源提供商社区中仍然是一种相对稀缺的商品,因此我们使用商业云提供商提供的GPU扩展了可用资源池。调配的资源通过开放科学网格(OSG)基础设施完全集成到正常的IceCube工作负载管理系统中,并使用CloudBank进行预算管理。其结果是,IceCube在两周内使用的GPU墙时数大约翻了一番,增加了超过3.1 fp32 EFLOP时数,价格约为5.8万美元。本文介绍了使用的设置和操作经验。

英文摘要:

The IceCube collaboration relies on GPU compute for many of its needs, including ray tracing simulation and machine learning activities. GPUs are however still a relatively scarce commodity in the scientific resource provider community, so we expanded the available resource pool with GPUs provisioned from the commercial Cloud providers. The provisioned resources were fully integrated into the normal IceCube workload management system through the Open Science Grid (OSG) infrastructure and used CloudBank for budget management. The result was an approximate doubling of GPU wall hours used by IceCube over a period of 2 weeks, adding over 3.1 fp32 EFLOP hours for a price tag of about $58k. This paper describes the setup used and the operational experience.

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。

↑