Application Runtime Variability and Power Optimization for Exascale Computers

Porterfield, Allan; Fowler, Rob; Bhalachandra, Sridutt; Rountree, Barry; Deb, Diptorup; Lewis, Rob

doi:10.1145/2768405.2768408

Cited by 19 publications

(8 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They demonstrate substantial power savings on mini‐apps with minimal performance change using these system metrics. Porterfield et al present experiments on distributed memory applications, enforcing power limits from within MPI communication libraries when collective operations are occurring. This approach is capable of saving energy for large‐scale applications with minimal costs to execution time.…”

Section: Related Workmentioning

confidence: 99%

Investigating power capping toward energy‐efficient scientific applications

Haidar

Jagode

Vaccaro

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

Summary The emergence of power efficiency as a primary constraint in processor and system design poses new challenges concerning power and energy awareness for numerical libraries and scientific applications. Power consumption also plays a major role in the design of data centers, which may house petascale or exascale‐level computing systems. At these extreme scales, understanding and improving the energy efficiency of numerical libraries and their related applications becomes a crucial part of the successful implementation and operation of the computing system. In this paper, we study and investigate the practice of controlling a compute system's power usage, and we explore how different power caps affect the performance of numerical algorithms with different computational intensities. Further, we determine the impact, in terms of performance and energy usage, that these caps have on a system running scientific applications. This analysis will enable us to characterize the types of algorithms that benefit most from these power management schemes. Our experiments are performed using a set of representative kernels and several popular scientific benchmarks. We quantify a number of power and performance measurements and draw observations and conclusions that can be viewed as a roadmap to achieving energy efficiency in the design and execution of scientific algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Investigating power capping toward energy‐efficient scientific applications

Haidar

Jagode

Vaccaro

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Sarood et al demonstrate that adding additional low power nodes to an execution may improve performance compared to running fewer high power nodes [25]. Porterfield et al boost power on processors that fall behind due to hardware performance variability under a power cap [22]. Several poweraware schedulers for HPC have been proposed [5,27,24,19].…”

Section: Related Workmentioning

confidence: 99%

Early experiences with node-level power capping on the Cray XC40 platform

Pedretti

Olivier

Ferreira

et al. 2015

Proceedings of the 3rd International Workshop on Energy Efficient Supercomputing

View full text Add to dashboard Cite

Power consumption of extreme-scale supercomputers has become a key performance bottleneck. Yet current practices do not leverage power management opportunities, instead running at "maximum power". This is not sustainable. Future systems will need to manage power as a critical resource, directing it to where it has greatest benefit. Power capping is one mechanism for managing power budgets, however its behavior is not well understood. This paper presents an empirical evaluation of several key HPC workloads running under a power cap on a Cray XC40 system, and provides a comparison of this technique with p-state control, demonstrating the performance differences of each. These results show: 1.) Maximum performance requires ensuring the cap is not reached; 2.) Performance slowdown under a cap can be attributed to cascading delays which result in unsynchronized performance variability across nodes; and, 3.) Due to lag in reaction time, considerable time is spent operating above the set cap. This work provides a timely and much needed comparison of HPC application performance under a power cap and attempts to enable users and system administrators to understand how to best optimize application performance on power-constrained HPC systems.

show abstract

“…The performance stability of the application is one of the essential factors of HPC systems [9]. However, applications show run-to-run variations in performance owing to many components in a system [1], [10]- [15]. Among these inhibitors, OS noise should be resolved because it severely limits the scalability and performance of applications [1].…”

Section: Introductionmentioning

confidence: 99%

A Performance-Stable NUMA Management Scheme for Linux-Based HPC Systems

et al. 2021

View full text Add to dashboard Cite

Linux is becoming the de-facto standard operating system for today's high-performance computing (HPC) systems because it can satisfy the demands of many HPC systems for rich operating system (OS) features. However, owing to features intended for the general-purpose OS, Linux has many OS noise sources such as page faults or thread migrations that can result in the unstable performance of HPC application. Furthermore, in the case of the non-uniform memory access (NUMA) architecture, which has different memory access latencies to local and remote memory nodes, the performance stability of the application can be more exacerbated by the OS noise. In this paper, we address the OS noise caused by Linux in the NUMA architecture and propose a novel performance-stable NUMA management scheme called Stable-NUMA. Stable-NUMA comprises three techniques for improving performance stability: two-level thread clustering, state-based page placement, and selective page profiling. Our proposed Stable-NUMA scheme significantly alleviates OS noise and enhances the local memory access ratio of the NUMA system as compared to Linux. We implemented Stable-NUMA in Linux and experimented with various HPC workloads. The evaluation results demonstrated that Stable-NUMA outperforms Linux with and without its NUMA-aware feature by up to 25% in terms of average performance and 73% in terms of performance stability.

show abstract

Application Runtime Variability and Power Optimization for Exascale Computers

Cited by 19 publications

References 38 publications

Investigating power capping toward energy‐efficient scientific applications

Investigating power capping toward energy‐efficient scientific applications

Early experiences with node-level power capping on the Cray XC40 platform

A Performance-Stable NUMA Management Scheme for Linux-Based HPC Systems

Contact Info

Product

Resources

About