NSDF-Cloud: Enabling Ad-Hoc Compute Clusters Across Academic and Commercial Clouds

NSDF-Cloud: Enabling Ad-Hoc Compute Clusters Across Academic and Commercial Clouds
复制标题

NSDF-Cloud:跨学术和商业云启用临时计算集群

DOI:
10.1145/3502181.3533710
复制
发表时间:
2022
期刊:
HPDC '22: Proceedings of the 31st International Symposium on High-Performance Parallel and Distributed Computing
影响因子:
--
通讯作者:
Taufer, Michela
Taufer, Michela
中科院分区:
--
文献类型:
--
作者:
Luettgau, Jakob;Olaya, Paula;Zhou, Naweiluo;Scorzelli, Giorgio;Pascucci, Valerio;Taufer, Michela

文献摘要

被引文献

相似文献

计算资源越来越多地通过类似云的界面提供给用户。学术和商业云产品都存在,但不存在用于配置、启动和终止虚拟资源等常见操作的单一标准化接口。这给试图利用这些资源的领域科学家带来了巨大的技术负担;即使是专家用户也会花费大量时间将他们的应用程序从一个云平台移植到另一个云平台。通过这项工作,我们向社区提供了一个统一的 API 工具包,以及关于我们与不同学术和商业云提供商合作时遇到的挑战的五份深入报告。我们的工具包实现了常见任务的自动化,例如跨云同时启动和终止大量虚拟机 (VM)。我们证明,我们的工具包将用户启动和终止这些资源所需的时间缩短到几分钟,从而实现临时多云集群。
Computational resources are increasingly provisioned to users through cloud-like interfaces. Both academic and commercial cloud offerings exist, but no single standardized interface for common actions such as configuration, launching, and termination of virtual resources exists. This imposes huge technical burden on domain scientist that attempt to take advantage of these resources; even expert users spend considerable time to port their applications from one cloud platform to another.With this work, we make available to the community a unified API toolkit as well as five in-depth reports on challenges we encountered working with different academic and commercial cloud providers. Our toolkit implements automations for common tasks such as simultaneous launching and termination of large numbers of virtual machines (VM) across the cloud. We demonstrate that our toolkit brings down the time users need to spend launching and terminating these resources to mere minutes, thus enabling ad-hoc multi-cloud clusters.