Demystifying asynchronous I/O Interference in HPC applications
Demystifying asynchronous I/O Interference in HPC applications
复制标题
揭秘 HPC 应用中的异步 I/O 干扰
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Aparna Chandramowlishwaran
中科院分区：
文献类型：
--
作者：
S. Tseng;Bogdan Nicolae;F. Cappello;Aparna Chandramowlishwaran
随着HPC工作流复杂性的增加，数据管理服务需要在后台异步执行昂贵的I/O操作，目的是将I/O与应用程序运行时重叠。但是，这可能会由于资源竞争而导致干扰：CPU，内存/网络带宽。多核架构的出现加剧了这个问题，因为许多I/O操作是并发发出的，因此不仅与应用程序竞争，而且在它们之间竞争。此外，干扰模式可以动态地改变，以响应应用行为和I/O子系统（例如，多个用户共享并行文件系统）的变化。如果没有彻底的理解，I/O操作可能会执行得不太理想，甚至可能比阻塞情况更糟。为了填补这一空白，本文调查的原因和后果的干扰，由于HPC系统上的异步I/O。具体来说，我们专注于多核CPU和内存带宽，隔离每个资源的干扰。然后，我们进行了深入的研究，以解释在各种资源共享的情况下，如不同的优先级和后台I/O线程的数量和不同的I/O策略：sendfile，读/写，mmap/写强调权衡的相互作用和竞争。这项研究的见解是重要的，既可以引导优化现有的后台I/O，以及打开新的机会，设计先进的异步I/O策略。
With increasing complexity of HPC workflows, data management services need to perform expensive I/O operations asynchronously in the background, aiming to overlap the I/O with the application runtime. However, this may cause interference due to competition for resources: CPU, memory/network bandwidth. The advent of multi-core architectures has exacerbated this problem, as many I/O operations are issued concurrently, thereby competing not only with the application but also among themselves. Furthermore, the interference patterns can dynamically change as a response to variations in application behavior and I/O subsystems (e.g. multiple users sharing a parallel file system). Without a thorough understanding, I/O operations may perform suboptimally, potentially even worse than in the blocking case. To fill this gap, this paper investigates the causes and consequences of interference due to asynchronous I/O on HPC systems. Specifically, we focus on multi-core CPUs and memory bandwidth, isolating the interference due to each resource. Then, we perform an in-depth study to explain the interplay and contention in a variety of resource sharing scenarios such as varying priority and number of background I/O threads and different I/O strategies: sendfile, read/write, mmap/write underlining trade-offs. The insights from this study are important both to enable guided optimizations of existing background I/O, as well as to open new opportunities to design advanced asynchronous I/O strategies.
挤压网格上有限元的结构利用编号算法及其在 Firedrake 中的性能评估
影响因子：
5.1
作者：
Bercea G
通讯作者：
Bercea G