Single processing units have now reached a point where further major improvements in their performance are restricted by their physical limitations. This is causing a slowing down in advances at the same time as new scientific challenges are demanding exascale speed. This has meant that parallel processing has become key to High Performance Computing (HPC). This book contains the proceedings of the 14th biennial ParCo conference, ParCo2011, held in Ghent, Belgium. The ParCo conferences have traditionally concentrated on three main themes: Algorithms, Architectures and Applications. Nowadays though, the focus has shifted from traditional multiprocessor topologies to heterogeneous and manycores, incorporating standard CPUs, GPUs (Graphics Processing Units) and FPGAs (Field Programmable Gate Arrays). These platforms are, at a higher abstraction level, integrated in clusters, grids and clouds. The papers presented here reflect this change of focus. New architectures, programming tools and techniques are also explored, and the need for exascale hardware and software was also discussed in the industrial session of the conference.This book will be of interest to all those interested in parallel computing today, and progress towards the exascale computing of tomorrow.
This book constitutes the thoroughly refereed post-conference proceedings of the 9th International Conference on High Performance Computing for Computational Science, VECPAR 2010, held in Berkeley, CA, USA, in June 2010. The 34 revised full papers presented together with five invited contributions were carefully selected during two rounds of reviewing and revision. The papers are organized in topical sections on linear algebra and solvers on emerging architectures, large-scale simulations, parallel and distributed computing, numerical algorithms.
Single processing units have now reached a point where further major improvements in their performance are restricted by their physical limitations. This is causing a slowing down in advances at the same time as new scientific challenges are demanding exascale speed. This has meant that parallel processing has become key to High Performance Computing (HPC).This book contains the proceedings of the 14th biennial ParCo conference, ParCo2011, held in Ghent, Belgium. The ParCo conferences have traditionally concentrated on three main themes: Algorithms, Architectures and Applications. Nowadays though, the focus has shifted from traditional multiprocessor topologies to heterogeneous and manycores, incorporating standard CPUs, GPUs (Graphics Processing Units) and FPGAs (Field Programmable Gate Arrays). These platforms are, at a higher abstraction level, integrated in clusters, grids and clouds. The papers presented here reflect this change of focus. New architectures, programming tools and techniques are also explored, and the need for exascale hardware and software was also discussed in the industrial session of the conference.This book will be of interest to all those interested in parallel computing today, and progress towards the exascale computing of tomorrow.
This open access book summarizes the research done and results obtained in the second funding phase of the Priority Program 1648 "Software for Exascale Computing" (SPPEXA) of the German Research Foundation (DFG) presented at the SPPEXA Symposium in Dresden during October 21-23, 2019. In that respect, it both represents a continuation of Vol. 113 in Springer’s series Lecture Notes in Computational Science and Engineering, the corresponding report of SPPEXA’s first funding phase, and provides an overview of SPPEXA’s contributions towards exascale computing in today's sumpercomputer technology. The individual chapters address one or more of the research directions (1) computational algorithms, (2) system software, (3) application software, (4) data management and exploration, (5) programming, and (6) software tools. The book has an interdisciplinary appeal: scholars from computational sub-fields in computer science, mathematics, physics, or engineering will find it of particular interest.
From the Foreword: "The authors of the chapters in this book are the pioneers who will explore the exascale frontier. The path forward will not be easy... These authors, along with their colleagues who will produce these powerful computer systems will, with dedication and determination, overcome the scalability problem, discover the new algorithms needed to achieve exascale performance for the broad range of applications that they represent, and create the new tools needed to support the development of scalable and portable science and engineering applications. Although the focus is on exascale computers, the benefits will permeate all of science and engineering because the technologies developed for the exascale computers of tomorrow will also power the petascale servers and terascale workstations of tomorrow. These affordable computing capabilities will empower scientists and engineers everywhere." — Thom H. Dunning, Jr., Pacific Northwest National Laboratory and University of Washington, Seattle, Washington, USA "This comprehensive summary of applications targeting Exascale at the three DoE labs is a must read." — Rio Yokota, Tokyo Institute of Technology, Tokyo, Japan "Numerical simulation is now a need in many fields of science, technology, and industry. The complexity of the simulated systems coupled with the massive use of data makes HPC essential to move towards predictive simulations. Advances in computer architecture have so far permitted scientific advances, but at the cost of continually adapting algorithms and applications. The next technological breakthroughs force us to rethink the applications by taking energy consumption into account. These profound modifications require not only anticipation and sharing but also a paradigm shift in application design to ensure the sustainability of developments by guaranteeing a certain independence of the applications to the profound modifications of the architectures: it is the passage from optimal performance to the portability of performance. It is the challenge of this book to demonstrate by example the approach that one can adopt for the development of applications offering performance portability in spite of the profound changes of the computing architectures." — Christophe Calvin, CEA, Fundamental Research Division, Saclay, France "Three editors, one from each of the High Performance Computer Centers at Lawrence Berkeley, Argonne, and Oak Ridge National Laboratories, have compiled a very useful set of chapters aimed at describing software developments for the next generation exa-scale computers. Such a book is needed for scientists and engineers to see where the field is going and how they will be able to exploit such architectures for their own work. The book will also benefit students as it provides insights into how to develop software for such computer architectures. Overall, this book fills an important need in showing how to design and implement algorithms for exa-scale architectures which are heterogeneous and have unique memory systems. The book discusses issues with developing user codes for these architectures and how to address these issues including actual coding examples.’ — Dr. David A. Dixon, Robert Ramsay Chair, The University of Alabama, Tuscaloosa, Alabama, USA
The research and its outcomes presented in this collection focus on various aspects of high-performance computing (HPC) software and its development which is confronted with various challenges as today's supercomputer technology heads towards exascale computing. The individual chapters address one or more of the research directions (1) computational algorithms, (2) system software, (3) application software, (4) data management and exploration, (5) programming, and (6) software tools. The collection thereby highlights pioneering research findings as well as innovative concepts in exascale software development that have been conducted under the umbrella of the priority programme "Software for Exascale Computing" (SPPEXA) of the German Research Foundation (DFG) and that have been presented at the SPPEXA Symposium, Jan 25-27 2016, in Munich. The book has an interdisciplinary appeal: scholars from computational sub-fields in computer science, mathematics, physics, or engineering will find it of particular interest.
In the last decade, parallel computing technologies have transformed high-performance computing. Two trends have emerged massively parallel computing leading to exascale on the one hand and moderately parallel applications, which have opened up high-perf
HPC, Big Data, AI Convergence Towards Exascale provides an updated vision on the most advanced computing, storage, and interconnection technologies, that are at basis of convergence among the HPC, Cloud, Big Data, and artificial intelligence (AI) domains. Through the presentation of the solutions devised within recently founded H2020 European projects, this book provides an insight on challenges faced by integrating such technologies and in achieving performance and energy efficiency targets towards the exascale level. Emphasis is given to innovative ways of provisioning and managing resources, as well as monitoring their usage. Industrial and scientific use cases give to the reader practical examples of the needs for a cross-domain convergence. All the chapters in this book pave the road to new generation of technologies, support their development and, in addition, verify them on real-world problems. The readers will find this book useful because it provides an overview of currently available technologies that fit with the concept of unified Cloud-HPC-Big Data-AI applications and presents examples of their actual use in scientific and industrial applications.
If you look around you will find that all computer systems, from your portable devices to the strongest supercomputers, are heterogeneous in nature. The most obvious heterogeneity is the existence of computing nodes of different capabilities (e.g. multicore, GPUs, FPGAs, ...). But there are also other heterogeneity factors that exist in computing systems, like the memory system components, interconnection, etc. The main reason for these different types of heterogeneity is to have good performance with power efficiency. Heterogeneous computing results in both challenges and opportunities. This book discusses both. It shows that we need to deal with these challenges at all levels of the computing stack: from algorithms all the way to process technology. We discuss the topic of heterogeneous computing from different angles: hardware challenges, current hardware state-of-the-art, software issues, how to make the best use of the current heterogeneous systems, and what lies ahead. The aim of this book is to introduce the big picture of heterogeneous computing. Whether you are a hardware designer or a software developer, you need to know how the pieces of the puzzle fit together. The main goal is to bring researchers and engineers to the forefront of the research frontier in the new era that started a few years ago and is expected to continue for decades. We believe that academics, researchers, practitioners, and students will benefit from this book and will be prepared to tackle the big wave of heterogeneous computing that is here to stay.
