A5035259771- Parallel Computing and Optimization Techniques
- Distributed and Parallel Computing Systems
- Advanced Data Storage Technologies
- Cloud Computing and Resource Management
- Scientific Computing and Data Management
- Software System Performance and Reliability
- Embedded Systems Design Techniques
- Teaching and Learning Programming
- VLSI and Analog Circuit Testing
- Particle Detector Development and Performance
- Big Data and Business Intelligence
- Medical Imaging Techniques and Applications
- Radiation Effects in Electronics
- Research, Science, and Academia
- Educational Leadership and Innovation
- SARS-CoV-2 and COVID-19 Research
- Software Testing and Debugging Techniques
- COVID-19 epidemiological studies
- Innovation, Sustainability, Human-Machine Systems
- Biomedical and Engineering Education
- Algorithms and Data Compression
- COVID-19 diagnosis using AI
- COVID-19 Pandemic Impacts
- Online Learning and Analytics
- Radiation Detection and Scintillator Technologies
Oak Ridge National Laboratory
2015-2024
Office of Scientific and Technical Information
2024
Oak Ridge Leadership Computing Facility
2023
Helmholtz-Zentrum Dresden-Rossendorf
2022
University of Illinois Urbana-Champaign
2022
University of Delaware
2022
Institute of Technology of Cambodia
2022
University of Basel
2022
Swisscom (Switzerland)
2022
Sandia National Laboratories
2022
CORAL, the Collaboration of Oak Ridge, Argonne and Livermore, is fielding two similar IBM systems, Summit Sierra, with NVIDIA GPUs that will replace existing Titan Sequoia systems. Sierra are currently ranked No. 1 3, respectively on Top500 list. We discuss design key differences Our evaluation systems highlights following. Applications fit in HBM see most benefit may prefer more GPUs; however, for some applications, CPU-GPU bandwidth important than number GPUs. The node-local burst buffer...
Summary The Oak Ridge Leadership Computing Facility (OLCF) has a long history of supporting and promoting GPU‐accelerated computing starting with the deployment Titan supercomputer in 2021 continuing Summit which theoretical peak performance approximately 200 petaflops. Because majority Summit's computational power comes from its 27,972 GPUs, users must port their applications to one supported programming models order make efficient use system. To prepare transition Frontier, OLCF's exascale...
The I/O subsystem for the Summit supercomputer, No. 1 on Top500 list, and its ecosystem of analysis platforms is composed two distinct layers, namely in-system layer center-wide parallel file system (PFS), Spider 3. uses node-local SSDs provides 26.7 TB/s reads, 9.7 writes, 4.6 billion IOPS to Summit. 3 PFS IBM's Spectrum Scale™ 2.5 2.6 million other systems. While deploying them as layers was operationally efficient, it also presented usability challenges in terms multiple mount points lack...
National labs, academic institutions and industry have a strong need for scientists staff that understand high performance computing (HPC) the complex interconnections across individual topics in HPC. However, domain science computer undergraduate programs are not providing sufficient educational resources, far from conveying interdisciplinary collaborative nature of HPC environment. The Student Cluster Competition (SCC) was created as an tool to immerse undergraduates It is microcosm...
Modern High Performance Computing (HPC) sys-tems are built with innovative system architectures and novel programming models to further push the speed limit of computing. The increased complexity poses challenges for performance portability evaluation. Standard Perfor-mance Evaluation Corporation (SPEC) has a long history producing industry-standard benchmarks modern computer systems. SPEC's newly released SPEChpc 2021 benchmark suites, developed by Group, bold attempt provide fair objective...
In this paper we explore the performance portability of directives provided by OpenMP 4 and OpenACC to program various types node architectures with attached accelerators, both self-hosted multicore offload multicore/GPU. Our goal is examine how successful newer features 4.5 are for moving codes between architectures, much tuning might be required what lessons can learn from experience. To do this, use examples algorithms varying computational intensities our evaluation, as compute data...
The SPEChpc 2021 suites are application-based benchmarks de- signed to measure performance of modern HPC systems. bench- marks support MPI, MPI+OpenMP, MPI+OpenMP target offload, MPI+OpenACC and portable across all major platforms.
In this paper we explore the performance portability of directives provided by OpenMP 4 and OpenACC to program various types node architectures with attached accelerators, both self-hosted multicore offload multicore/GPU. Our goal is examine how successful newer features 4.5 are for moving codes between architectures, much tuning might be required what lessons can learn from experience. To do this, use examples algorithms varying computational intensities our evaluation, as compute data...
Summary This article studies mini‐applications—Minisweep, GenASiS , GPP, and FF—that use computational methods commonly encountered in HPC. We have ported these applications to develop OpenACC OpenMP versions, evaluated their performance on Titan (Cray XK7 with K20x GPUs), Cori XC40 Intel KNL), Summit (IBM AC922 Volta Cori‐GPU CS‐Storm 500NX Skylake GPUs). Our goals are for new ports be useful both application compiler developers, document describe the lessons learned methodology create...
Summary The molecular dynamics simulation software, LAMMPS, utilizes the Kokkos acceleration library to port computation a diverse set of architectures including those based on GPU accelerators. In addition Kokkos, LAMMPS contains vast code base that leverages CUDA application programming interface using functions such as cuFFT , CUDA's fast‐fourier transform (FFT) library, and, more recently, also support for AMD's Heterogeneous Interface Portability (HIP) is rapidly growing. While...
This paper assesses and reports the experience of ten teams working to port, validate, benchmark several High Performance Computing applications on a novel GPU-accelerated Arm testbed system. The consists eight NVIDIA HPC Developer Kit systems, each one equipped with server-class CPU from Ampere two data center GPUs Corp. systems are connected together using InfiniBand interconnect. selected mini-apps written programming languages use multiple accelerator-based models for such as CUDA,...
Summary The Oak Ridge Leadership Computing Facility (OLCF) has been preparing the nation's first exascale system, Frontier, for production and end users. Frontier is based on HPE Cray's new EX architecture Slingshot interconnect features 74 cabinets of optimized 3rd Gen AMD EPYC CPUs HPC AI Instinct 250X accelerators. As a part this preparation, “real‐world” user codes have selected to help assess functionality, performance, usability system. This article describes early experiences using...
Summary We measure and analyze the performance observed when running applications benchmarks before after Meltdown Spectre fixes have been applied to Cray supercomputers supporting systems at Oak Ridge Leadership Computing Facility (OLCF). Of particular interest is effect of these on selected from OLCF portfolio scale. This comprehensive study presents results experiments run Titan, Eos, Cumulus, Percival OLCF. The this are useful for HPC users serve better understand impact that two...
In May 2022, the newest supercomputer to top TOP 500 list was Frontier at Oak Ridge National Laboratory, demonstrating capability of computing more than 1.1 quintillion (1018) floating-point calculations every second. Driving this ground-breaking rate is Frontier's 37,000 graphics processing units (GPUs) and 9,408 central (CPUs). total, contains 60 million parts. At scale, smallest margin error may generate hundreds hardware errors across system. These are capable directly hindering...