Filters








757 Hits in 3.7 sec

D5.2: Best Practices for HPC Procurement and Infrastructure

Norbert Meyer, Marcin Lawenda
2013 Zenodo  
Task 1 – Assessment of petascale systems – has performed a continuous market watch and analysis of trends in petascale HPC systems worldwide.  ...  Specific areas of interest are analysed in depth in terms of the market they belong to and the general HPC landscape, with a particular emphasis on the European point of view.  ...  The current supercomputer is a cluster with a peak performance of 94 Tflop/s. It is installed in a mobile data center and based on a container solution provided by Bull (MOBULL).  ... 
doi:10.5281/zenodo.6572412 fatcat:2bqftmr5zzb7na6pnlrlxxuqnu

D5.1: Preliminary Guidance on Procurements and Infrastructure

Guillermo Aguirre
2013 Zenodo  
Task 1 – Assessment of petascale systems – has performed a continuous market watch and analysis of trends in petascale HPC systems worldwide.  ...  Specific areas of interest are analysed in depth in terms of the market they belong to and the general HPC landscape, with a particular emphasis on the European point of view.  ...  Peak performance As per the definition of this list of petascale systems, all have a peak performance of at least 1 PFlop/s.  ... 
doi:10.5281/zenodo.6572385 fatcat:zyqybkd6dbhpzoikwus3mgwtme

D8.4: Final Report on Petascale Systems and Centre Assessment

Guillermo Aguirre, François Robin, Jean-Philippe Nominé, Marco Sbrighi
2012 Zenodo  
Task 1 - Assessment of petascale systems - has performed a continuous market watch and analysis of trends.  ...  Task 2 - Design and operation of reliable and power efficient HPC computer centres - has produced a se [...]  ...  be taken to try to measure or predict the performance of an application on a given system.  ... 
doi:10.5281/zenodo.6553025 fatcat:n6vcl4z4srapvl2usgepnvoqje

D5.3: Updated Best Practices for HPC Procurement and Infrastructure

Andreas Johansson
2014 Zenodo  
Task 1 – Assessment of petascale systems – has performed a continuous market watch and analysis of trends in petascale HPC systems worldwide.  ...  Specific areas of interest are analysed in depth in terms of the market they belong to and the general HPC landscape, with a particular emphasis on the European point of view.  ...  This cannot be done by just scaling currently available codes, but by implementing a new paradigm for the development of advanced numerical methods to really exploit the intrinsic capabilities of the future  ... 
doi:10.5281/zenodo.6572433 fatcat:cwiqrgf33jajjjvhubky4f6rau

Science and Engineering in the Petascale Era

Thom H. Dunning, Klaus Schulten, Jeroen Tromp, Jeremiah P. Ostriker, Kelvin Droegemeier, Ming Xue, Paul Fussell
2009 Computing in science & engineering (Print)  
Petascale computing will enable more accurate predictions of protein structures; higher fidelity simulation of systems that represent major units of life itself (such as viruses and ribosomes); and more  ...  Such systems entail a dramatic increase in computing capability that will, in nearly all science and engineering fields, create breakthrough advances, including • Predictive models of biological systems  ...  Ming Xue ("Atmospheric Science") is a professor of meteorology at the University of Oklahoma, where he is director of the Center of Analysis and Prediction of Storms.  ... 
doi:10.1109/mcse.2009.145 pmid:21998556 pmcid:PMC3190305 fatcat:rbaljddsirgeva5e3gwozgjvae

Robust Machine Learning Applied to Terascale Astronomical Datasets [article]

Nicholas M. Ball, Robert J. Brunner, Adam D. Myers (University of Illinois at Urbana-Champaign, National Center for Supercomputing Applications)
2007 arXiv   pre-print
Through a combination of machine learning in the form of decision trees, k-nearest neighbor, and genetic algorithms, the use of supercomputing resources at NCSA, and the cyberenvironment Data-to-Knowledge  ...  , we are able to provide improved classifications for over 100 million objects in the SDSS, improved photometric redshifts, and a full exploitation of the powerful k-nearest neighbor algorithm.  ...  Each node is a Dell PowerEdge 1750 server running Red Hat Linux with two Intel Xeon 3.2 GHz processors, 3 GB of memory, a peak double-precision performance of 6.4 Gflops, and 70 GB of scratch disk space  ... 
arXiv:0710.4482v1 fatcat:yjsajqyvf5cwjhk6b3xs4jdn6i

Visualization for Exascale: Portable Performance is Critical

2015 Supercomputing Frontiers and Innovations  
We conclude by reviewing a comparison of a visualization algorithm written with data parallel primitives and separate versions hand written for different architectures to show comparable performance with  ...  We can achieve performance portability by optimizing this small set of data parallel primitives whose tuning conveys to the conglomerates.  ...  a petascale supercomputer to an expected exascale supercomputer [1] Exascale (Prediction) Visualization for Exascale: Portable Performance is Critical  ... 
doi:10.14529/jsfi150306 fatcat:epdzc3xqvbh75f5t7ucoxndc5a

Robust Machine Learning Applied to Terascale Astronomical Datasets [article]

Nicholas M. Ball National Center for Supercomputing Applications, Urbana-Champaign)
2008 arXiv   pre-print
This is a novel application in the field of astronomy, because we are using such resources for data mining, and not just performing simulations.  ...  A driving principle of this work is that our methods should be extensible from current terascale datasets to upcoming petascale datasets and beyond.  ...  The authors made extensive use of the storage and computing facilities at the National Center for Supercomputing Applications and thank the technical staff for their assistance in enabling this work.  ... 
arXiv:0804.3417v1 fatcat:lxncos26ebfcxdi3g3ei447bwy

Energy-Efficient Virtual Screening with ARM-CPU-Based Computers

Olivia Alford, David Toth
2017 The Journal of Computational Science Education  
We tested a cluster of ODROID-XU3 Lite computers with ARM CPUs and compared its performance to a server with x64 CPUs.  ...  In order to be a viable solution, our cluster needed to perform the screen without sacrificing speed or increasing hardware costs.  ...  We used a slightly customized version of AutoDock Vina that outputs only the best score, rather than all the data that Vina outputs by default, to minimize file I/O [3] .  ... 
doi:10.22369/issn.2153-4136/8/2/3 fatcat:n763crgg6jbt7oigx27d4te6eu

Core Concept: Nascent exascale supercomputers offer promise, present challenges

Adam Mann
2020 Proceedings of the National Academy of Sciences of the United States of America  
In 2008, when Los Alamos Laboratory unveiled Roadrunner, the world's first petascale supercomputer, it contained 12,960 GPU-inspired chips along with 6,480 CPUs and performed twice as well as the next  ...  GPUs were a boon for supercomputers.  ... 
doi:10.1073/pnas.2015968117 pmid:32907934 fatcat:j6ucqopzvvcbnmn3oksgjookne

D3.1.7 Proceedings of second scientific conference

Anni Jakobsson, Naomi Messing-Klopstra
2009 Zenodo  
The theme of the symposium was "HPC Infrastructures for Petascale Applications", it was of major interest to a broad audience: from scientific users, HPC technology experts and vendors to government representatives  ...  one big European HPC (High Performance Computing) event: The DEISA PRACE Symposium 2009.  ...  The occurrence of a well documented secondary peak of velocity fluctuations due to adverse pressure gradient is the consequences of a very strong production.  ... 
doi:10.5281/zenodo.6543209 fatcat:avlcreat2zg2lp2flcwkdx2lwq

Extreme-scale scripting: Opportunities for large task-parallel applications on petascale computers

Michael Wilde, Ioan Raicu, Allan Espinosa, Zhao Zhang, Ben Clifford, Mihael Hategan, Sarah Kenny, Kamil Iskra, Pete Beckman, Ian Foster
2009 Journal of Physics, Conference Series  
Parallel scripting is a loosely-coupled programming model in which applications are composed of highly parallel scripts of program invocations that process and exchange data via files.  ...  with parallel scripts on currently available petascale computers.  ...  Each is capable of consuming a large fraction, or even all, of a petascale computer.  ... 
doi:10.1088/1742-6596/180/1/012046 fatcat:w6hdr5mluvgy5lt53bk5llpvu4

The quest for petascale computing

J.J. Dongarra, D.W. Walker
2001 Computing in science & engineering (Print)  
This article discusses the current status of ef-forts to make petascale computing a reality, but first we review trends in supercomputer performance over the past 50 years.  ...  A corollary of this is that for certain problems, it will be impossible to predict a priori which solution method will perform best or converge at all.  ... 
doi:10.1109/5992.919263 fatcat:mrvgy6ytcra3bhqoul67ffor4e

Operating system issues for petascale systems

Pete Beckman, Kamil Iskra, Kazutomo Yoshii, Susan Coghlan
2006 ACM SIGOPS Operating Systems Review  
Petascale supercomputers will be available by 2008. The largest machine of these complex leadership-class machines will probably have nearly 250K CPUs.  ...  These massively parallel systems have a number of challenging operating system issues.  ...  All of these studies show that the core issue is synchronicity of operating systems for petascale machines. PETASCALE I/O Data input/output can be a considerable problem on petascale machines.  ... 
doi:10.1145/1131322.1131332 fatcat:e6whyflugncn3bdeneycuhawvm

Performance Impact of I/O on QMCPack Simulations at the Petascale and Beyond

S. Herbein, M. Matheny, M. Wezowicz, J. Krogel, J. Logan, J. Kim, S. Klasky, M. Taufer
2013 2013 IEEE 16th International Conference on Computational Science and Engineering  
The extension of a tool such as Skel to mimic the variable I/O in QMCPack allows us to predict the I/O performance of the code when using ADIOS methods at the petascale.  ...  Traditional petascale applications, such as QMC-Pack, can scale their computations to completely utilize modern supercomputers like Titan, but they cannot scale their I/O.  ...  It was also partially supported by the projects Predictive Theory and Modeling for Materials and Chemical Science by the Department of Energy (DOE), Basic Energy Science (BES), the Laboratory Directed  ... 
doi:10.1109/cse.2013.24 dblp:conf/cse/HerbeinMWKLKKT13 fatcat:i3s25ou5yvbejowvcionu3b57a
« Previous Showing results 1 — 15 out of 757 results