发布求助
文献互助 智能选刊 最新文献

2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)最新文献

筛选
英文 中文
SPOTlight on testing: Stability, performance and operational testing of LANL HPC clusters 测试焦点:LANL高性能计算集群的稳定性、性能和运行测试
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063348.2063382
G. Pedicini, Jennifer Green
{"title":"SPOTlight on testing: Stability, performance and operational testing of LANL HPC clusters","authors":"G. Pedicini, Jennifer Green","doi":"10.1145/2063348.2063382","DOIUrl":"https://doi.org/10.1145/2063348.2063382","url":null,"abstract":"Testing is sometimes a forgotten component of system management, but it becomes very important in the realm of High Performance Computing (HPC) clusters. Many large-scale HPC cluster installations are one of a kind, with unknown issues and unexpected behaviors. First, the initial installation may uncover complex configuration interactions that are only apparent at scale; Stability becomes a critical feature of early system testing. Second, Performance may be significantly impacted by small changes to the system. Third, after initial shakeout, users expect a system that is reliable on their terms; ongoing Operational tests verify reliability, and provide early warning of developing problems. A robust test suite should address all of these test categories, and present both tests and results in a manner that meets usability requirements. We will describe Los Alamos National Laboratory's current test suite, and the development project to expand the suite to cover these areas and provide better tools for analysis and reporting.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127011978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
A fast solver for modeling the evolution of virus populations 模拟病毒种群进化的快速求解器
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063384.2063483
Gerhard Niederbrucker, W. Gansterer
{"title":"A fast solver for modeling the evolution of virus populations","authors":"Gerhard Niederbrucker, W. Gansterer","doi":"10.1145/2063384.2063483","DOIUrl":"https://doi.org/10.1145/2063384.2063483","url":null,"abstract":"Solving Eigen's quasispecies model for the evolution of virus populations involves the computation of the dominant eigen vector of a matrix whose size N grows exponentially with the chain length of the virus to be modeled. Most biologically interesting chain lengths are so far well beyond the reach of existing algorithms and hardware. We show how to exploit the special properties of the problem under consideration and design a fast and accurate solver which reduces the complexity to Θ(N log2 N). Our solver is even faster than existing approximative strategies and contrary to those it can also be applied to more general formulations of the quasispecies model. Substantial further improvements and high parallelism can be achieved for special fitness landscapes in the evolution model. Beyond theoretical analysis, we evaluate the performance of our new solver experimentally on a GPU with an OpenCL implementation and illustrate that it achieves speedup factors of more than 107 over standard approaches.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"269 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125825600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Simplified parallel domain traversal 简化的并行域遍历
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063384.2063397
W. Kendall, Jingyuan Wang, M. Allen, T. Peterka, Jian Huang, David Erickson
{"title":"Simplified parallel domain traversal","authors":"W. Kendall, Jingyuan Wang, M. Allen, T. Peterka, Jian Huang, David Erickson","doi":"10.1145/2063384.2063397","DOIUrl":"https://doi.org/10.1145/2063384.2063397","url":null,"abstract":"Many data-intensive scientific analysis techniques require global domain traversal, which over the years has been a bottleneck for efficient parallelization across distributed- memory architectures. Inspired by MapReduce and other simplified parallel programming approaches, we have designed DStep, a flexible system that greatly simplifies efficient parallelization of domain traversal techniques at scale. In order to deliver both simplicity to users as well as scalability on HPC platforms, we introduce a novel two-tiered communication architecture for managing and exploiting asynchronous communication loads. We also integrate our design with advanced parallel I/O techniques that operate directly on native simulation output. We demonstrate DStep by performing teleconnection analysis across ensemble runs of terascale atmospheric CO2 and climate data, and we show scalability results on up to 65,536 IBM BlueGene/P cores.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126569795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 51
Performance modeling for systematic performance tuning 用于系统性能调优的性能建模
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063348.2063356
T. Hoefler, W. Gropp, W. Kramer, M. Snir
{"title":"Performance modeling for systematic performance tuning","authors":"T. Hoefler, W. Gropp, W. Kramer, M. Snir","doi":"10.1145/2063348.2063356","DOIUrl":"https://doi.org/10.1145/2063348.2063356","url":null,"abstract":"The performance of parallel scientific applications depends on many factors which are determined by the execution environment and the parallel application. Especially on large parallel systems, it is too expensive to explore the solution space with series of experiments. Deriving analytical models for applications and platforms allow estimating and extrapolating their execution performance, bottlenecks, and the potential impact of optimization options. We propose to use such \"performance modeling\" techniques beginning from the application design process throughout the whole software development cycle and also during the lifetime of supercomputer systems. Such models help to guide supercomputer system design and re-engineering efforts to adopt applications to changing platforms and allow users to estimate costs to solve a particular problem. Models can often be built with the help of well-known performance profiling tools. We discuss how we successfully used modeling throughout the proposal, initial testing, and beginning deployment phase of the Blue Waters supercomputer system.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124889244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 95
Fast implementation of DGEMM on Fermi GPU DGEMM在Fermi GPU上的快速实现
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063384.2063431
Guangming Tan, Linchuan Li, Sean Triechle, E. Phillips, Yungang Bao, Ninghui Sun
{"title":"Fast implementation of DGEMM on Fermi GPU","authors":"Guangming Tan, Linchuan Li, Sean Triechle, E. Phillips, Yungang Bao, Ninghui Sun","doi":"10.1145/2063384.2063431","DOIUrl":"https://doi.org/10.1145/2063384.2063431","url":null,"abstract":"In this paper we present a thorough experience on tuning double-precision matrix-matrix multiplication (DGEM-M) on the Fermi GPU architecture. We choose an optimal algorithm with blocking in both shared memory and registers to satisfy the constraints of the Fermi memory hierarchy. Our optimization strategy is further guided by a performance modeling based on micro-architecture benchmarks. Our optimizations include software pipelining, use of vector memory operations, and instruction scheduling. Our best CUDA algorithm achieves comparable performance with the latest CUBLAS library1. We further improve upon this with an implementation in the native machine language, leading to 20% increase in performance. That is, the achieved peak performance (efficiency) is improved from 302Gflop/s (58%) to 362Gflop/s (70%).","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123692682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 119
Deep and wide metrics for HPC resource capability and project usage 对高性能计算资源能力和项目使用情况进行深入和广泛的度量
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063348.2063350
David L. Hart
{"title":"Deep and wide metrics for HPC resource capability and project usage","authors":"David L. Hart","doi":"10.1145/2063348.2063350","DOIUrl":"https://doi.org/10.1145/2063348.2063350","url":null,"abstract":"This paper defines and demonstrates application of possible quantitative metrics for the qualitative notions of \"deep\" and \"wide\" HPC system use along with the related concepts of capability and capacity computing. By summarizing HPC workloads according to the science-oriented projects using the systems, rather than solely on job sizes and distributions, one can identify differences between the workloads on different systems as well as highlight certain instances of unique usage modalities. Specific definitions of depth and width are suggested, along with metrics that permit comparisons to determine which systems are deeper or wider than others. Similarly, the same data permit an alternate means of defining the degree to which HPC system activities are capability- or capacity-oriented.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"99 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134541895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 8
Hardware/software co-design for energy-efficient seismic modeling 节能地震建模的软硬件协同设计
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063384.2063482
J. Krueger, D. Donofrio, J. Shalf, M. Mohiyuddin, Samuel Williams, L. Oliker, F. Pfreundt
{"title":"Hardware/software co-design for energy-efficient seismic modeling","authors":"J. Krueger, D. Donofrio, J. Shalf, M. Mohiyuddin, Samuel Williams, L. Oliker, F. Pfreundt","doi":"10.1145/2063384.2063482","DOIUrl":"https://doi.org/10.1145/2063384.2063482","url":null,"abstract":"Reverse Time Migration (RTM) has become the standard for high-quality imaging in the seismic industry. RTM relies on PDE solutions using stencils that are 8th order or larger, which require large-scale HPC clusters to meet the computational demands. However, the rising power consumption of conventional cluster technology has prompted investigation of architectural alternatives that offer higher computational efficiency. In this work, we compare the performance and energy efficiency of three architectural alternatives the Intel Nehalem X5530 multicore processor, the NVIDIA Tesla C2050 GPU, and a general-purpose manycore chip design optimized for high-order wave equations called \"Green Wave.\" We have developed an FPGA-accelerated architectural simulation platform to accurately model the power and performance of the Green Wave design. Results show that across highly-tuned high-order RTM stencils, the Green Wave implementation can offer up to 8× and 3.5× energy efficiency improvement per node respectively, compared with the Nehalem and GPU platforms. These results point to the enormous potential energy advantages of our hardware/software co-design methodology.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133535066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 45
Performance of the Community Earth System Model 社区地球系统模型的性能
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063384.2063457
P. Worley, A. Mirin, A. Craig, M. Taylor, J. Dennis, M. Vertenstein
{"title":"Performance of the Community Earth System Model","authors":"P. Worley, A. Mirin, A. Craig, M. Taylor, J. Dennis, M. Vertenstein","doi":"10.1145/2063384.2063457","DOIUrl":"https://doi.org/10.1145/2063384.2063457","url":null,"abstract":"The Community Earth System Model (CESM), released in June 2010, incorporates new physical process and new numerical algorithm options, significantly enhancing simulation capabilities over its predecessor, the June 2004 release of the Community Climate System Model. CESM also includes enhanced performance tuning options and performance portability capabilities. This paper describes performance and performance scaling on both the Cray XT5 and the IBM BG/P for four representative production simulations, varying both problem size and enabled physical processes. The paper also describes preliminary performance results for high resolution simulations using over 200,000 processor cores, indicating the promise of ongoing work in numerical algorithms and where further work is required.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129416017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 38
“Hot” for warm water cooling “Hot”表示温水冷却
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063348.2063371
Henry Coles, M. Ellsworth, David J. Martinez
{"title":"“Hot” for warm water cooling","authors":"Henry Coles, M. Ellsworth, David J. Martinez","doi":"10.1145/2063348.2063371","DOIUrl":"https://doi.org/10.1145/2063348.2063371","url":null,"abstract":"Liquid cooling is key to reducing energy consumption for this generation of supercomputers and remains on the roadmap for the foreseeable future. This is because the heat capacity of liquids is orders of magnitude larger than that of air and once heat has been transferred to a liquid, it can be removed from the datacenter efficiently. The transition from air to liquid cooling is an inflection point providing an opportunity to work collectively to set guidelines for facilitating the energy efficiency of liquid-cooled High Performance Computing (HPC) facilities and systems. The vision is to use non-compressor-based cooling, to facilitate heat re-use, and thereby build solutions that are more energy-efficient, less carbon intensive and more cost effective than their air-cooled predecessors. The Energy Efficient HPC Working Group is developing guidelines for warmer liquid-cooling temperatures in order to standardize facility and HPC equipment, and provide more opportunity for reuse of waste heat. This report describes the development of those guidelines.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132783672","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 15
Qserv: A distributed shared-nothing database for the LSST catalog Qserv:用于LSST编目的分布式无共享数据库
2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC) Pub Date : 2011-11-12 DOI: 10.1145/2063348.2063364
Daniel L. Wang, S. Monkewitz, Kian-Tat Lim, J. Becla
{"title":"Qserv: A distributed shared-nothing database for the LSST catalog","authors":"Daniel L. Wang, S. Monkewitz, Kian-Tat Lim, J. Becla","doi":"10.1145/2063348.2063364","DOIUrl":"https://doi.org/10.1145/2063348.2063364","url":null,"abstract":"The LSST project will provide public access to a database catalog that, in its final year, is estimated to include 26 billion stars and galaxies in dozens of trillion detections in multiple petabytes. Because we are not aware of an existing open-source database implementation that has been demonstrated to efficiently satisfy astronomers' spatial self-joining and cross-matching queries at this scale, we have implemented Qserv, a distributed shared-nothing SQL database query system. To speed development, Qserv relies on two successful open-source software packages: the MySQL RDBMS and the Xrootd distributed file system. We describe Qserv's design, architecture, and ability to scale to LSST's data requirements. We illustrate its potential with test results on a 150-node cluster using 55 billion rows and 30 terabytes of simulated data. These results demonstrate the soundness of Qserv's approach and the scale it achieves on today's hardware.","PeriodicalId":358797,"journal":{"name":"2011 International Conference for High Performance Computing, Networking, Storage and Analysis (SC)","volume":"189 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116609610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 22
首页 上一页 下一页 尾页
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整，为了账户安全，请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们：info@booksci.cn Book学术提供免费学术资源搜索服务，方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号：481959085
Book学术官方微信