Proceedings International Parallel and Distributed Processing Symposium最新文献_第2页

Supporting the hard real-time requirements of mechatronic systems by 2-level interrupt service management 通过二级中断服务管理，支持机电一体化系统的硬实时性要求

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213236

Christian Siemers, R. Falsett, R. Seyer, K. Ecker

引用次数: 3

A new DMA registration strategy for pinning-based high performance networks 一种新的基于钉接的高性能网络DMA注册策略

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213363

Christian Bell, D. Bonachea

{"title":"A new DMA registration strategy for pinning-based high performance networks","authors":"Christian Bell, D. Bonachea","doi":"10.1109/IPDPS.2003.1213363","DOIUrl":"https://doi.org/10.1109/IPDPS.2003.1213363","url":null,"abstract":"This paper proposes anew memory registration strategy for supporting Remote DMA (RDMA) operations over pinning-based networks, as existing approaches are insufficient for efficiently implementing Global Address Space (GAS) languages. Although existing approaches often maximize bandwidth, they require levels of synchronization that discourage one-sided communication, and can have significant latency costs for small messages. The proposed Firehose algorithm attempts to expose one-sided, zero-copy communication as a common case, while minimizing the number of host-level synchronizations required to support remote memory operations. The basic idea is to reap the performance benefits of a pin-everything approach in the common case (without the drawbacks) and revert to a rendezvous-based approach to handle the uncommon case. In all cases, the algorithm attempts to amortize the cost of synchronization and pinning over multiple remote memory operations, improving performance over rendezvous by avoiding many handshaking messages and the cost of re-pinning recently used pages. Performance results are presented which demonstrate that the cost of two-sided handshaking and memory registration is negligible when the set of remotely referenced memory pages on a given node is smaller than the physical memory (where the entire working set can remain pinned), and for applications with larger working sets the performance degrades gracefully and consistently outperforms conventional approaches.","PeriodicalId":177848,"journal":{"name":"Proceedings International Parallel and Distributed Processing Symposium","volume":"263 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116037566","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: 117

Approximation in non-product form multiple queue systems 非乘积形式多队列系统的近似

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213507

N. Thomas

引用次数: 1

Performance monitoring and evaluation of a UPC implementation on a NUMA architecture 在NUMA架构上UPC实现的性能监控和评估

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213492

François Cantonnet, Yiyi Yao, Smita Annareddy, A. Mohamed, T. El-Ghazawi

引用次数: 25

The unlinkability of randomization-enhanced Chaum's blind signature scheme 随机增强Chaum盲签名方案的不可链接性

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213443

Zichen Li, Junmei Zhang, W. Kou

引用次数: 2

New dynamic heuristics in the client-agent-server model 客户机-代理-服务器模型中的新动态启发式

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213200

Y. Caniou, E. Jeannot

引用次数: 7

MUSE: a software oscilloscope for clusters and grids MUSE:用于集群和网格的软件示波器

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213096

M. Gardner, M. Broxton, Adam Engelhart, Wu-chun Feng

{"title":"MUSE: a software oscilloscope for clusters and grids","authors":"M. Gardner, M. Broxton, Adam Engelhart, Wu-chun Feng","doi":"10.1109/IPDPS.2003.1213096","DOIUrl":"https://doi.org/10.1109/IPDPS.2003.1213096","url":null,"abstract":"Oscilloscopes and their cousins, logic analyzers, are the tools of choice for difficult electronic hardware problems. In the hands of a skilled engineer or technician, these tools can be used to solve stubborn problems. The key to the utility of oscilloscopes is the depth of detail they provide and their flexibility, which allows the level of detail to be adjusted to fit the task at hand. Distributed applications, which run on computing clusters and computational grids, are also complex and difficult to tame. We need tools to understand their complexities and the ability to choose the level of detail to fit the task, whether the task be debugging, tuning, monitoring or controlling. The MAGNET User-Space Environment (MUSE) has been designed as a \"software oscilloscope\" for computing clusters and computational grids. It is a toolkit for applications and developers to obtain detailed information about the environment on the host. The information can be used on-line or saved for off-line analysis. It has low overhead and allows the level of detail to be adjusted. Furthermore, MUSE monitors without requiring the modification or relinking of applications. It has been designed to make it easy to develop \"adaptive applications\" - applications that are aware of their environment and can adapt to changes.","PeriodicalId":177848,"journal":{"name":"Proceedings International Parallel and Distributed Processing Symposium","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128328729","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

Improved methods for divisible load distribution on k-dimensional meshes using pipelined communications 基于流水线通信的k维网格可分负载分配改进方法

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213185

Keqin Li

引用次数: 13

Energy-aware compilation and execution in Java-enabled mobile devices 支持java的移动设备中的能量感知编译和执行

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213116

Guilin Chen, B. Kang, M. Kandemir, N. Vijaykrishnan, M. J. Irwin, R. Chandramouli

{"title":"Energy-aware compilation and execution in Java-enabled mobile devices","authors":"Guilin Chen, B. Kang, M. Kandemir, N. Vijaykrishnan, M. J. Irwin, R. Chandramouli","doi":"10.1109/IPDPS.2003.1213116","DOIUrl":"https://doi.org/10.1109/IPDPS.2003.1213116","url":null,"abstract":"Java-enabled wireless devices are preferred for various reasons such as enhanced user experience and the support for dynamically downloading applications on demand. The dynamic download capability supports extensibility of the mobile client features and centralizes application maintenance at the server. Also, it enables service providers to customize features for the clients. In this work, we extend this client-server collaboration further by offloading some of the computations (i.e., method execution and dynamic compilation) normally performed by the mobile client to the resource-rich server in order to conserve energy consumed by the client in a wireless Java environment. In the proposed framework, the object serialization feature of Java is used to allow offloading of both method execution and bytecode-to-native code compilation to the server when executing a Java application. Our framework takes into account communication, computation and compilation energies to dynamically decide where to compile and execute a method (locally or remotely) and how to execute it (using interpretation or just-in-time compilation with different levels of optimizations).","PeriodicalId":177848,"journal":{"name":"Proceedings International Parallel and Distributed Processing Symposium","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129002823","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: 10

Continuous compilation: a new approach to aggressive and adaptive code transformation 持续编译:一种积极和自适应代码转换的新方法

Proceedings International Parallel and Distributed Processing Symposium Pub Date : 2003-04-22 DOI: 10.1109/IPDPS.2003.1213375

B. Childers, J. Davidson, M. Soffa

{"title":"Continuous compilation: a new approach to aggressive and adaptive code transformation","authors":"B. Childers, J. Davidson, M. Soffa","doi":"10.1109/IPDPS.2003.1213375","DOIUrl":"https://doi.org/10.1109/IPDPS.2003.1213375","url":null,"abstract":"Over the past several decades, the compiler research community has developed a number of sophisticated and powerful algorithms for a variety of code improvements. While there are still promising directions for particular optimizations, research on new or improved optimizations is reaching the point of diminishing returns and new approaches are needed to achieve significant performance improvements beyond traditional optimizations. In this paper, we describe a new strategy based on a continuous compilation system that constantly improves application code by applying aggressive and adaptive code optimizations at all times, from static optimization to online dynamic optimization. In this paper, we describe our general approach and process for continuous compilation of application code. We also present initial results from our research with continuous compilation. These initial results include a new prediction framework that can estimate the benefit of applying code transformations without actually doing the transformation. We also describe results that demonstrate the benefit of adaptively changing application code for embedded systems to make trade-offs between code size, performance, and power consumption.","PeriodicalId":177848,"journal":{"name":"Proceedings International Parallel and Distributed Processing Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129132750","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