首页> 外文会议>Proceedings of the Twentieth annual symposium on Parallelism in algorithms and architectures >Directed transmission method, a fully asynchronous approach to solve sparse linear systems in parallel
【24h】

Directed transmission method, a fully asynchronous approach to solve sparse linear systems in parallel

机译:定向传输方法,一种完全异步的方法,用于并行解决稀疏线性系统

获取原文
获取原文并翻译 | 示例

摘要

This volume consists of papers that were presented at the 20th ACM Symposium on Parallelism in Algorithms and Architectures (SPAA'07), held on June 14-16, 2008, in Munich, Germany. It was sponsored by the ACM Special Interest Groups on Algorithms and Computation Theory (SIGACT) and Computer Architecture (SIGARCH) and organized in cooperation with the European Association for Theoretical Computer Science (EATCS). Financial support was provided by the German Research Foundation (DFG), Akamai, Intel, and Sun Microsystems. >The 36 regular presentations that appeared at the conference were selected by the program committee after an electronic discussion and a meeting of two days that took place at Sun Labs, Burlington, Massachusetts, on March 20-21. Of these papers, the paper "Utilizing Shared Data in Chip Multiprocessors with the Nahalal Architecture" by Zvika Guz, Idit Keidar, Avinoam Kolodny, and Uri Weiser was selected to receive the best paper award. Fourteen papers were selected by Jim Larus for the special track on "Hardware and Software Techniques for Multicore Programming". >The regular presentations were selected out of 128 submitted abstracts. The mix of selected papers reflects the unique nature of SPAA in bringing together the theory and practice of parallel computing. SPAA defines parallelism very broadly to encompass any computational device or scheme that can perform multiple operations or tasks simultaneously or concurrently. This year, the selection of papers was closer to SPAA's roots, in that technical papers included a large number of papers on parallel and multicore algorithms and architectures. The technical papers in this volume are to be considered preliminary versions, and authors are generally expected to publish polished and complete versions in archival scientific journals. >In addition to the regular presentations, this volume includes 13 brief announcements and the abstracts of 4 posters that were presented at the SPAA review sessions. The committee's decisions in accepting brief announcements and posters were based on the perceived interest of these contributions, with the goal that they serve as bases for further significant advances in parallelism in computing. Extended versions of the SPAA brief announcements and posters may be published later in other conferences or journals. >Finally, this year, there were 6 invited talks by Kunle Olukotun, Arnold Rosenberg, Leslie Valiant, Hagit Attiya, Charles Leiserson, and Burton Smith, to celebrate SPAA's 20 th anniversary.
机译:本卷包含论文,论文于2008年6月14日至16日在德国慕尼黑举行的第20届ACM算法和体系结构并行论研讨会(SPAA'07)上发表。它由ACM算法和计算理论特别兴趣小组(SIGACT)和计算机体系结构(SIGARCH)赞助,并与欧洲理论计算机科学协会(EATCS)合作组织。财务支持由德国研究基金会(DFG),Akamai,英特尔和Sun Microsystems提供。

会议上出现的36篇常规演讲是由程序委员会在电子讨论和会议之后选定的。 3月20日至21日在马萨诸塞州伯灵顿的Sun Labs举行了为期两天的会议。在这些论文中,由Zvika Guz,Idit Keidar,Avinoam Kolodny和Uri Weiser撰写的论文“在具有Nahalal架构的芯片多处理器中利用共享数据”获得了最佳论文奖。 Jim Larus选择了14篇论文作为“多核编程的硬件和软件技术”的专题。

从128篇提交的摘要中选择了常规演示文稿。精选论文的混合反映了SPAA在汇集并行计算的理论和实践方面的独特性质。 SPAA非常广泛地定义了并行性,以涵盖可以同时或同时执行多个操作或任务的任何计算设备或方案。今年,论文的选择更接近SPAA的根源,因为技术论文包括大量有关并行和多核算法和体系结构的论文。本册子中的技术论文应被视为初稿,通常应期望作者在档案科学期刊上发表完整的完整版。

除常规介绍外,本册子还包括13条简短的公告和在SPAA审查会议上展示的4张海报的摘要。该委员会接受简短公告和海报的决定是基于这些贡献的可察觉兴趣,其目的是作为它们在计算并行性方面取得更大进步的基础。 SPAA简要公告和海报的扩展版本可能会在以后的其他会议或期刊上发表。 Leiserson和Burton Smith共同庆祝SPAA成立20周年。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号