Filters








3,092 Hits in 8.3 sec

Evaluating use of data flow systems for large graph analysis

Andy Yoo, Ian Kaplan
2009 Proceedings of the 2nd Workshop on Many-Task Computing on Grids and Supercomputers - MTAGS '09  
910 nodes (4 dual-core processors, 4 disks, 8 GB memory) Hadoop Map/Reduce Performed TB sort on 20-node DAS system Lawrence Livermore National Laboratory Achieved 2X speedup by • Radix-based distribution  ...  Uni-and Bi-directional BFS Pattern Matching Find subgraphs that matches given template Applications evaluated on DAS System 15 Lawrence Livermore National Laboratory template TeraByte (TB  ...  Path Traversal: Breadth-first search (BFS) on DAS Sou rce Destin ation Improved performance by constructing adjacent list via denormalization, which reduces the number of rows to join  ... 
doi:10.1145/1646468.1646473 dblp:conf/sc/YooK09 fatcat:jhvqsxx62zfyleh6b2t42xqqme

The high performance storage system

R. A. Coyne, H. Hulen, R. Watson
1993 Proceedings of the 1993 ACM/IEEE conference on Supercomputing - Supercomputing '93  
The NSL was organized to develop, storage system and data management requirements, HPSS demonstrate, and commercialize high-performance is designed using network-connected storage devices to storage system  ...  There are many requirements and challenges motivating the need for a new generation of storage The National Storage Laboratory (NSL) was organized system such as HPSS [4,5].  ...  It is anticipated that by the time of shared design, implementation, and operation experiences the second release, that should be toward the end of 1994 with our colleagues in the National Storage Laboratory  ... 
doi:10.1145/169627.169662 dblp:conf/sc/CoyneHW93 fatcat:zc75wkisbzemjdw5mprtempv5i

Utilizing Lustre file system with dCache for CMS analysis

Y Wu, B Kim, J L Rodriguez, Y Fu, D Bourilkov, P Avery
2010 Journal of Physics, Conference Series  
Our initial CMS analysis job measurement and transfer performance results are shown and the advantages of difference implementations are briefly discussed.  ...  Our initial CMS analysis job measurement and transfer performance results are shown and the advantages of difference implementations are briefly discussed. Abstract.  ...  Acknowledgments Authors wishing to acknowledge a lot of assistances and advices from the colleagues at UF HPC Centre, especially Dr. Craig Prescott.  ... 
doi:10.1088/1742-6596/219/6/062068 fatcat:vbtzuajb6fcf7dak6isejmklay

Scaling NFSv4 with parallel file systems

D. Hildebrand, P. Honeyman
2005 CCGrid 2005. IEEE International Symposium on Cluster Computing and the Grid, 2005.  
This paper introduces extensions to the NFSv4 protocol to support parallel access. We implemented a prototype of our design and present experiments demonstrating its scalable architecture.  ...  This paper introduces extensions to the NFSv4 protocol to support parallel access. We implemented a prototype of our design and present experiments demonstrating its scalable architecture.  ...  This research is partially supported by Lawrence Livermore National Laboratory, Los Alamos National Laboratory, and Sandia National Laboratories under contract B523296.  ... 
doi:10.1109/ccgrid.2005.1558675 dblp:conf/ccgrid/HildebrandH05 fatcat:b462cdqgjvhhvobgccpev44fju

THE HIGH PERFORMANCE STORAGE SYSTEM (HPSS)

K. L. KLIEWER
1996 Computing in High Energy Physics '95  
To help meet these requirements, the scalable, network-centered, parallel storage system HPSS was designed and is now being developed.  ...  The current development status and the broad applicability are illustrated through a discussion of the sites at which HPSS is now being implemented, representing a spectrum of computing environments.  ...  Optimum structure of the COS Hints will emerge only through experience and a careful assessment of effective storage system usage from both users and system administrators.  ... 
doi:10.1142/9789814447188_0014 fatcat:63zue6mqvzdcfmz2th463s2osu

Status of the NIF power conditioning system

P. A. Arnold, S.D. Hulsey, G.T. Ullery, D.E. Petersen, D.L. Pendleton, C.W. Ollis, M.A. Newton, T. Harwell, D. Cordoza, L. Hadovski
2007 2007 16th IEEE International Pulsed Power Conference  
The NIF Power Conditioning System provides the pulsed excitation required to drive flashlamps in the laser's optical amplifiers.  ...  Modular in design, each of the 192 Main Energy Storage Modules (MESMs) storage up to 2.2 MJ of electrical energy in its capacitor bank before delivering the energy to 20 pairs of flashlamps in a 400 µs  ...  Conventional facilities work has ended and the main emphasis has shifted to "filling the building" with hardware, completing the implementation and testing of the software and making the entire facility  ... 
doi:10.1109/ppps.2007.4652462 fatcat:43vofjqmgjantpz2h22nagkbmm

Status of the NIF Power Conditioning System

S.D. Hulsey, G.T. Ullery, D.E. Petersen, D.L. Pendleton, C.W. Ollis, M.A. Newton, P. A. Arnold, T. Harwell, D. Cordoza
2007 2007 IEEE Pulsed Power Plasma Science Conference  
The NIF Power Conditioning System provides the pulsed excitation required to drive flashlamps in the laser's optical amplifiers.  ...  Modular in design, each of the 192 Main Energy Storage Modules (MESMs) storage up to 2.2 MJ of electrical energy in its capacitor bank before delivering the energy to 20 pairs of flashlamps in a 400 µs  ...  Conventional facilities work has ended and the main emphasis has shifted to "filling the building" with hardware, completing the implementation and testing of the software and making the entire facility  ... 
doi:10.1109/ppps.2007.4346163 fatcat:netd4cx2tnajfjxpent5ms6asi

High Performance Storage System Scalability: Architecture, Implementation and Experience

R.W. Watson
22nd IEEE / 13th NASA Goddard Conference on Mass Storage Systems and Technologies (MSST'05)  
The High Performance Storage System (HPSS) provides scalable hierarchical storage management (HSM), archive, and file system services.  ...  This paper discusses HPSS architectural, implementation and deployment experiences that contributed to its success in meeting the above orders of magnitude scaling targets.  ...  This work was, in part, performed by the Lawrence Livermore National Laboratory, Los Alamos National Laboratory, Oak Ridge National Laboratory, National Energy Research Supercomputer Center and Sandia  ... 
doi:10.1109/msst.2005.17 dblp:conf/mss/Watson05 fatcat:fi2sbwghcfao3o4qrtewtuefwe

Novette Pulse Power System Description

D. G. Gritton, D. J. Christie, R. W. Holloway, B. T. Merritt, J. A. Oicles, K. Whitham, R. B. Wilcox
1983 Nuclear Technology - Fusion  
The Novette laser projected per formance exceeds that of the 20-arm Shiva laser and the frequency-multiplying capability provides more-efficient target interaction.  ...  Novette is a large, two-beam, two-wavelength laser facility at Lawrence Livermore National Laboratory.  ...  We. attribute this to our design of the facility grounding system, to the use of coaxial cables for distribution of pulsed power to laser amplifiers, and to the use of fiber optics for isolation.  ... 
doi:10.13182/fst83-a22982 fatcat:w2l2hd36hfgiznhbilcvm6g6zm

Characterization and Optimization of Memory-Resident MapReduce on HPC Systems

Yandong Wang, Robin Goldstone, Weikuan Yu, Teng Wang
2014 2014 IEEE 28th International Parallel and Distributed Processing Symposium  
Using a state-of-the-art MapReduce implementation Spark and the Hyperion system at Lawrence Livermore National Laboratory, we have examined the impact of storage architectures, data locality and task scheduling  ...  This work characterizes the performance impact of key differences between compute-and data-centric paradigms and then provides optimizations to enable a dual-purpose HPC system that can efficiently support  ...  This work was also performed under the auspices of the US Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 (LLNL-CONF-647813).  ... 
doi:10.1109/ipdps.2014.87 dblp:conf/ipps/WangGYW14 fatcat:b65tfnopz5d4tpkkn5olzudski

The IEEE Mass Storage System Reference Model [chapter]

SHIERS JAMIE
1994 Zenodo  
The model identifies the high level abstractions that underlie modern storage systems. The model itself does not attempt to provide implementation specifications.  ...  The IEEE Reference Model for Mass Storage Systems provides a basis for the develop­ment of standards for storage systems.  ...  The objectives of the workshop were to assemble a group of mass storage specialists whose task was to document ideas and experience from existing systems and to present these at the 6th IEEE Mass Storage  ... 
doi:10.5281/zenodo.4139255 fatcat:xg6za2x4abdkbcr4c3siafxwne

Data layout optimization for petascale file systems

Xian-He Sun, Yong Chen, Yanlong Yin
2009 Proceedings of the 4th Annual Workshop on Petascale Data Storage - PDSW '09  
Implementation results under MPI-IO middleware and PVFS2 file system confirm the correctness and effectiveness of their approach, and demonstrate the potential of data layout optimization in petascale  ...  In this study, the authors propose a simple performance model to promote a better integration between the parallel I/O middleware layer and parallel file systems.  ...  ACKNOWLEDGMENTS The authors are thankful to Dr. Rajeev Thakur, Dr. Rob Ross and Sam Lang of Argonne National Laboratory for their constructive and thoughtful suggestions toward this study.  ... 
doi:10.1145/1713072.1713077 fatcat:2uitwmakafc5ditqa45jpsviiq

Data acquisition system for Alcator C-Mod

T. W. Fredian, J. A. Stillerman, M. Greenwald
1997 Review of Scientific Instruments  
SUMMARY The data acquisition and analysis system implemented for Alcator C-Mod records and distributes the experimental data in an efficient and timely manner.  ...  This article will describe the computer hardware and software used to collect, store, and distribute the information generated by the Alcator C-Mod experiment. II.  ... 
doi:10.1063/1.1147775 fatcat:calpe3ekqjfdrp755v7xl4vtsa

Providing Quality of Service Support in Object-Based File System

Joel C. Wu, Scott A. Brandt
2007 24th IEEE Conference on Mass Storage Systems and Technologies (MSST 2007)  
Ceph is a highly scalable distributed file system that can scale up to tens of thousands of object-based storage devices (OSDs).  ...  Bourbon is a quality of service framework designed to work with the Ceph object-based storage system.  ...  We would also like to thank Lawrence Livermore, Los Alamos, and Sandia National Laboratories; and the UCSC/Los Alamos Institute for Scalable Scientific Data Management (ISSDM) for support.  ... 
doi:10.1109/msst.2007.4367971 fatcat:bqzrpvkb3ve6pfkcmevmrrzqwu

Hierarchical disk sharing for multimedia systems

Joel C. Wu, Scott Banachowski, Scott A. Brandt
2005 Proceedings of the international workshop on Network and operating systems support for digital audio and video - NOSSDAV '05  
We discuss the design of HDS and present our implementation in a Linux block device driver, demonstrating the effectiveness (and limitations) of this approach.  ...  Systems that use or serve multimedia data require timely access to data on hard drives.  ...  Acknowledgments This research was supported in part by Lawrence Livermore National Laboratory, Los Alamos National Laboratory, and Sandia National Laboratory under contract B520714; also by Intel Corporation  ... 
doi:10.1145/1065983.1066026 dblp:conf/nossdav/WuBB05 fatcat:jaoqdr2mtvgohn4w3mvaebqut4
« Previous Showing results 1 — 15 out of 3,092 results