Vetter Contemporary High Performance Computing

From Petascale toward Exascale, Volume Two
Erscheinungsjahr 2018
ISBN: 978-1-4987-0063-4
Verlag: CRC Press
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 255 Seiten

Reihe: Chapman & Hall/CRC Computational Science

HPC is used to solve a number of complex questions in computational and data-intensive sciences. These questions include the simulation and modeling of physical phenomena, such as climate change, energy production, drug design, global security, and materials design; the analysis of large data sets such as those in genome sequencing, astronomical observation, and cybersecurity; and the intricate design of engineered products, such as airplanes and automobiles.

This second volume of Contemporary High Performance Computing: From Petascale toward Exascale continues to document international HPC ecosystems, including the sponsors and sites that host them. Each chapter is punctuated with a site's flagship system and:

- Presents highlights of applications, workloads, and benchmarks

- Describes hardware architectures, system software, and programming systems

- Explores storage, visualization, and analytics

- Examines the data center/facility as well as system statistics

Featuring pictures of buildings and systems in production, floorplans, and many block diagrams and charts to illustrate system design and performance, Contemporary High Performance Computing: From Petascale toward Exascale, Volume Two delivers a detailed snapshot of the rich history of practice in modern HPC. This book provides a valuable reference for researchers in HPC and computational science.

Zielgruppe

Researchers in high performance computing and computational science.

Autoren/Hrsg.

Vetter, Jeffrey S.

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Rechnerarchitektur
• Mathematik | Informatik EDV | Informatik Technische Informatik Grid-Computing & Paralleles Rechnen

Inhaltsverzeichnis

Preface

List of Figures

List of Tables

Introduction; Jeffrey S. Vetter

Terascale to Petascale: The Past 17 Years in HPC

Bibliography

ARCHER; Xu Guo, David Henty, Chris Johnson, Alan Simpson, Lorna Smith, and Andre Turner

Overview

Sponsor/Program Background

Timeline

Applications and Workloads

Highlights of Main Applications

System Overview

Hardware Architecture

Compute Node Architecture

Interconnect

Service Node Architecture

External Login Nodes (esLogin)

Pre- and Post-Processing Nodes

Storage Systems

System Software

Operating System

Job Submission System: PBS Pro/ALPS

Programming System

Programming Models

Languages and Compilers

Performance Tools

Debugging

Long-Term Storage and Data Analytics

UK Research Data Facility Overview

UK-RDF Technical Details

Access to the UK-RDF and Moving Data

Data Analytic Workflows

Data Center/Facility

Location and History

Infrastructure

Innovations and Challenges

System Statistics

SAFE: Service Administration from EPCC

Bibliography

Enabling Extreme Scale Science at NERSC; Sudip Dosanjh, Jeff Broughton, Katie Antypas, Brent Draney, Jason Hick, Tina Declerck, and Nicholas Wright

Edison Supercomputer – NERSC-7

User Base and Science Areas

Procurement Strategy

Requirements Gathering

Sustained Performance

Best Value

Overlapping Systems

System Architecture

Processor and Memory

Interconnect

Cabinet and Chassis

Storage and I/O

System Software

Physical Infrastructure

Computational Research and Theory Facility

Cray XC30 Cooling Design

User Services

Early Application Results on Edison

Sequestered C02

Large-Scale Structure of the Universe

Graphene and Carbone Nanotubes

Better Combustion for New Fuels

Exascale Computing and the Future of NERSC

NERSC-8 as a Pre-Exascale System

Application Readiness

Conclusion

Acknowledgements

Bibliography

HLRN-III at Zuse Institute Berlin; Wolfgang Baumann, Guido Laubender, Matthias Läuter, Alexander Reinefeld, Christian Schimmel, Thomas Steinke, Christian Tuma, and Stefan Wollny

Introduction

Scientific Computing and Computer Science at ZIB

Supercomputers at ZIB: Past to Present

HLRN Supercomputing Alliance

Bodies

Funding

Procurements

Benchmarks or How We Found "Mrs. Perfect"

Scientific Domains and Workloads

Research Fields Exploiting the HLRN-III

Chemistry and Material Sciences

Earth Sciences

Engineering

Physics

HLRN-III System Architecture

Hardware Configuration

Software Ecosystem

System Software

Software for Program Development and Optimization

Application Software, Packages and Libraries

Storage Strategy and Pre- and Post-Processing

Data Center Facility

Accounting and User Management: Single System View over Two Sites

Preparing for the Future

Second Phase Installation of HLRN-III

Many-Core in HPC

HLRN-IV and Beyond

Bibliography

The K Computer; Mitsuo Yokokawa, Fumiyoshi Shoji, and Yukihiro Hasegawa

Overview

Development Targets and Schedule

Application Software Development Programs

System Overview

Hardware

System Software

Programming Models

Summary

Early Results

Benchmark Results

Gordon Bell Prizes

Operation

Early Access prior to Official Operation

Operation Policy after Official Operation

Utilization Statistics

Benchmark Applications

NICAM and Seism3D

FFB

LatticeQCD

PHASE and RSDFT

Optimized Performance

Facilities

Bibliography

Lindgren—The Swedish Tier-1 System; Erwin Laure, Daniel Ahlin, Genet Edmondson, Lars Malinowsky, Gert Svensson, and Jonathan Vincent

Overview

PDC Center for High-Performance Computing

Why Lindgren Was Needed

Lindgren Project Timeline

Applications and Workloads on Lindgren

Highlights of Lindgren's Main Applications

Overview of the Lindgren System

Lindgren's System Software

Lindgren's Programming System

Storage, Visualization, and Analytics on Lindgren

Data Center Facilities Housing Lindgren

Lindgren System Statistics

Lindgren Heat Re-Use System

The Future after Lindgren

Bibliography 167

Peregrine at the National Renewable Energy Laboratory; Kevin Regimbal, Ilene Carpenter, Christopher Chang, and Steve Hammond

Overview

Design Features, Efficiency & Sustainability Measures

Sponsor/Program Background

Timeline

Applications and Workloads

Computational Tasks and Domain Examples

Application Highlights

System Overview

Hardware Architecture

System Software

Programming System

Visualization and Analysis

Data Center/Facility

NREL Energy Efficient Data Center

ESIF Data Center Mechanical Infrastructure

ESIF Data Center Electrical Power

Power Interruptions

Fire Suppression

Physical Security

Return on Investment

System Statistics

Bibliography

Yellowstone: A Dedicated Resource for Earth System Science; Dr. Richard Loft, Aaron Andersen, Dr. Frank Bryan, Dr. John M. Dennis, Tom Engel, Pam Gillman, David Hart, Irian Elahi, Siddhartha Ghosh, Rory Kelly, Anke Kamrath, Dr. Gabrielle Pfister, Dr. Matthias Rempel, Dr. Justin Small, Dr. William Skamarock, Dr. Michael Wiltberger, Dr. Bryan Shader, Dr. Po Chen, and Dr. Ben Cash

Overview

Science Motivation

Sponsor and Program Background

Project Timeline

NWSC Construction

Yellowstone Procurement

System Overview

Hardware Architecture

Processors and Nodes

Rack-Level Packaging

Interconnect

Storage System

System Software

Operating Systems and System Management

Disk Filesystem and Tape Archive

System Monitoring

Programming Environment

Workload and Application Performance

Application Domain Descriptions

Application Performance

System Statistics

System Usage Patterns

Reliability, Uptime, and Utilization

Early Science Results

The NWSC Facility

Overview and Design

Early Experience in Operation

Future Challenges

Acknowledgments and Contributions

Bibliography

Über Autor(innen)

Jeffrey S. Vetter holds a joint appointment at Oak Ridge National Laboratory (ORNL) and the Georgia Institute of Technology (Georgia Tech). At ORNL, he is a distinguished R&D staff member and the founding group leader of the Future Technologies Group in the Computer Science and Mathematics Division. At Georgia Tech, Dr. Vetter is a professor in the Computational Science and Engineering School of the College of Computing, the principal investigator for the NSF XSEDE Keeneland GPU Supercomputer, and the director of the NVIDIA CUDA Center of Excellence. He earned his Ph.D in computer science from Georgia Tech. A senior member of the IEEE and a distinguished scientist member of the ACM, Dr. Vetter has published more than 130 peer-reviewed papers and has been a recipient of the ACM Gordon Bell Prize. His current research explores the role of emerging technologies in high performance computing.