Power limits on future high-performance computing (HPC) systems will constrain applications. However, HPC applications do not consume constant power over their lifetimes. Thus, applications assigned a fixed power bound may be forced to slow down during high-power computation phases, but may not consume their full power allocation during low-power I/O phases. This paper explores algorithms that leverage application semantics -- phase frequency, duration and power needs -- to shift unused power from applications in I/O phases to applications in computation phases, thus improving system-wide performance. We design novel techniques that include explicit staggering of applications to improve power shifting. Compared to executing without power shifting, our algorithms can improve average performance by up to 8% or improve performance of a single, high-priority application by up to 32%.
24 October 2019 in Journal of Open Source Software
(Filhol & Sturm, 2015; Kobayashi, 1980; Kochanski, Anderson, & Tucker, 2018). They change the reflectivity and average thermal conductivity of snow, and may change the patterns of snow accumulation and transport. Despite these effects, however, snow bedforms are poorly understood and not yet included in major snow or climate models.
Performance tuning of High-Performance Computing (HPC) applications depends on sophisticated tuning of parameters on diverse architectures. These parameters are made available by vendors through low-level dials such as model-specific registers (MSRs). While the MSRs themselves provide a powerful mechanism for users to monitor and control processor features, accessing them is laborious due to lack of standard interfaces and clear documentation. As a result, the burden of determining which MSRs to consider and how to fine-tune them for an application lies with the end user. In this paper, we present msr-genie, an efficient and extensible query tool that reduces this user-level burden. The tool allows them to query bi-directionally across MSR lists as well as processor families and models, and provides them with guidance on appropriate bitmasks. The msr-genie tool is open-source and easily extensible, and we demonstrate its effectiveness with over thirty Intel processor models and over two-thousand unique MSRs.