The Design, Deployment, and Evaluation of the CORAL Pre-Exascale Systems

Vazhkudai, Sudharshan S.; Supinski, Bronis R. de; Bland, Arthur S.; Geist, Al; Sexton, James; Kahle, J.; Zimmer, Christopher; Atchley, Scott; Oral, Sarp; Maxwell, Don; Vergara, Verónica Melesse; Bertsch, Adam; Goldstone, Robin; Joubert, Wayne; Chambreau, Chris; Appelhans, David; Blackmore, Robert; Casses, Ben; Chochia, George; Davison, Gene; Ezell, Matthew A; Gooding, Tom; Gonsiorowski, Elsa; Grinberg, Leopold; Hanson, Bill; Hartner, Bill; Karlin, Ian; Leininger, Matthew L.; Leverman, Dustin; Marroquin, Chris; Moody, Adam; Ohmacht, Martin; Pankajakshan, Ramesh; Pizzano, Fernando; Rogers, James H.; Rosenburg, Bryan S.; Schmidt, Drew; Shankar, Mallikarjun; Wang, Feiyi; Watson, Py C.; Walkup, Bob; Weems, Lance D.; Yin, Junqi

doi:10.1109/sc.2018.00055

Cited by 126 publications

(37 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…高性能低功耗的存储器是体系结构必须解决的问题. 异构计算系统大致可分为两类, 一类称之为节点内异构, 美国的 "TITAN", 中国的 "天河一号"、 "天河二号", 以及美国最新的 "Summit" [19] [23] , 由软件把高层业务逻辑转换成数据流逻辑, 然后按照数据流配置 FPGA 实现的硬件. 输入数据流入系统, 经过配置好的数据流硬件处理, 流出的就是结果.…”

Section: 面临的挑战unclassified

Key issues in exascale computing

Qian¹,

Wang²

2020

Sci. Sin.-Inf.

View full text Add to dashboard Cite

Over the past several decades, high performance computing (HPC) in China has undergone tremendous growth under the continuous support of national research programs. The development of exascale computers is the current goal set by the National Key R&D Project on HPC. Starting with a brief historical review of China's HPC development, this article analyzes the major challenges encountered in developing exascale computers. Thereafter, some important issues in realizing exascale computing are discussed, including architecture, processor, interconnect, parallel system software, parallel programming, algorithm, and resilience.

show abstract

Section: 面临的挑战unclassified

Key issues in exascale computing

Qian¹,

Wang²

2020

Sci. Sin.-Inf.

View full text Add to dashboard Cite

show abstract

“…Summit. As the No.1 supercomputer in the latest Top500 list (Dongarra et al 2018), Summit is designed for the workload of the United States Department of Energy (DoE) Office of Science and other domains such as natural science and engineering (Vazhkudai et al 2018).…”

Section: Analysis Of Pre-exascale I/o Architecturementioning

confidence: 99%

“…Shared burst buffer has a large chunk of global high-speed storage which can be accessed by all compute nodes (or several nodes share a burst buffer block). This kind of architecture can support the "N-1" I/O access pattern well (i.e., N processes writing to a single, shared file)) (Vazhkudai et al 2018). 2local burst buffer.…”

Section: Characteristics Of Pre-exascale and Projected Exascale I/o Amentioning

confidence: 99%

Software-defined QoS for I/O in exascale computing

et al. 2019

View full text Add to dashboard Cite

Supercomputers' capability is approaching the exascale level, which enables large computing systems to run more jobs concurrently. Since modern data-intensive scientific applications can sometimes produce millions of I/O requests per second, I/O systems always suffer from heavy workloads and impede the overall performance. How to allocate I/O resources and guarantee the QoS (Quality of Service) for each individual application is becoming an increasingly important question. In this paper, we propose SDQoS, a software-defined QoS framework with the token bucket algorithm, aiming to meet the I/O requirements of concurrent applications contending for the I/O resources and improve the overall performance of the I/O systems. Evaluation shows that SDQoS can effectively control the I/O bandwidth within a 5%-10% deviation and improve the performance by 20% in extreme cases.

show abstract

“…Since then, the increase of computing power is largely from an increased density of computing units in the processors. As a result, for the leading-edge supercomputers that were developed after 2010, a major part of computing power is provided by many-core accelerators, such as NVIDIA GPUs (Vazhkudai et al, 2018), Intel Xeon Phi MICs (Liao et al, 2014), and even reconfigurable FPGAs (Gan et al, 2017;Chen, 2019). As a result, recent machines contain two major architectural changes.…”

Section: Introductionmentioning

confidence: 99%

Optimizing High-Resolution Community Earth System Model on a Heterogeneous Many-Core Supercomputing Platform (CESM-HR_sw1.0)

Zhang¹,

Fu²,

Wu³

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract. With the semi-conductor technology gradually approaching its physical and heat limits, recent supercomputers have adopted major architectural changes to continue increasing the performance through more power-efficient heterogeneous many-core systems. Examples include Sunway TaihuLight that has four Management Processing Element (MPE) and 256 Computing Processing Element (CPE) inside one processor and Summit that has two central processing units (CPUs) and 6 graphics processing units (GPUs) inside one node. Meanwhile, current high-resolution Earth system models that desperately require more computing power, generally consist of millions of lines of legacy codes developed for traditional homogeneous multi-core processors and cannot automatically benefit from the advancement of supercomputer hardware. As a result, refactoring and optimizing the legacy models for new architectures become a key challenge along the road of taking advantage of greener and faster supercomputers, providing better support for the global climate research community and contributing to the long-lasting society task of addressing long-term climate change. This article reports the efforts of a large group in the International Laboratory for High-Resolution Earth System Prediction (iHESP) established by the cooperation of Qingdao Pilot National Laboratory for Marine Science and Technology (QNLM), Texas A & M University and the National Center for Atmospheric Research (NCAR), with the goal of enabling highly efficient simulations of the high-resolution (25-km atmosphere and 10-km ocean) Community Earth System Model (CESM-HR) on Sunway TaihuLight. The refactoring and optimizing efforts have improved the simulation speed of CESM-HR from 1 SYPD (simulation years per day) to 3.4 SYPD (with output disabled), and supported several hundred years of pre-industrial control simulations. With further strategies on deeper refactoring and optimizing for a few remaining computing hot spots, we expect an equivalent or even better efficiency than homogeneous CPU platforms. The refactoring and optimizing processes detailed in this paper on the Sunway system should have implications to similar efforts on other heterogeneous many-core systems such as GPU-based high-performance computing (HPC) systems.

show abstract

The Design, Deployment, and Evaluation of the CORAL Pre-Exascale Systems

Cited by 126 publications

References 9 publications

Key issues in exascale computing

Key issues in exascale computing

Software-defined QoS for I/O in exascale computing

Optimizing High-Resolution Community Earth System Model on a Heterogeneous Many-Core Supercomputing Platform (CESM-HR_sw1.0)

Contact Info

Product

Resources

About