International Conference on Parallel Processing, 2004. ICPP 2004.最新文献_第2页

The impact of MPI queue usage on message latency MPI队列使用对消息延迟的影响

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327915

K. Underwood, R. Brightwell

引用次数: 53

A genetic algorithm based approach for scheduling decomposable data grid applications 基于遗传算法的可分解数据网格应用调度方法

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327949

Seonho Kim, J. Weissman

{"title":"A genetic algorithm based approach for scheduling decomposable data grid applications","authors":"Seonho Kim, J. Weissman","doi":"10.1109/ICPP.2004.1327949","DOIUrl":"https://doi.org/10.1109/ICPP.2004.1327949","url":null,"abstract":"Data grid technology promises geographically distributed scientists to access and share physically distributed resources such as compute resource, networks, storage, and most importantly data collections for large-scale data intensive problems. Because of the massive size and distributed nature of these datasets, scheduling data grid applications must consider communication and computation simultaneously to achieve high performance. In many data grid applications, data can be decomposed into multiple independent sub datasets and distributed for parallel execution and analysis. We exploit this property and propose a novel genetic algorithm based approach that automatically decomposes data onto communication and computation resources. The proposed GA-based scheduler takes advantage of the parallelism of decomposable data grid applications to achieve the desired performance level. We evaluate the proposed approach comparing with other algorithms. Simulation results show that the proposed GA-based approach can be a competitive choice for scheduling large data grid applications in terms of both scheduling overhead and the relative solution quality as compared to other algorithms.","PeriodicalId":106240,"journal":{"name":"International Conference on Parallel Processing, 2004. ICPP 2004.","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125901267","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: 101

RGB: a scalable and reliable group membership protocol in mobile Internet RGB:移动互联网中可扩展、可靠的组成员协议

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327939

Guojun Wang, Jiannong Cao, Keith C. C. Chan

{"title":"RGB: a scalable and reliable group membership protocol in mobile Internet","authors":"Guojun Wang, Jiannong Cao, Keith C. C. Chan","doi":"10.1109/ICPP.2004.1327939","DOIUrl":"https://doi.org/10.1109/ICPP.2004.1327939","url":null,"abstract":"We propose a membership protocol for group communications in mobile Internet. The protocol is called RGB, which is the acronym of \"a ring-based hierarchy of access proxies, access gateways, and border routers\". RGB runs in a parallel and distributed way in the sense that each network entity in the ring-based hierarchy maintains local information about its possible leader, previous, next, parent and child neighbors, and that each network entity independently collects/generates membership change information, which is propagated by the one-round membership algorithm concurrently running in all the logical rings. We prove that the proposed protocol is scalable in the sense that the scalability of a ring-based hierarchy is as good as that of a tree-based hierarchy. We also prove that the proposed protocol is reliable, in the sense that, with high probability of 99.500%, a ring-based hierarchy with up to 1000 access proxies attached by a large number of mobile hosts will not partition when node faulty probability is bounded by 0.1%; if at most 3 partitions are allowed, then the function-well probability of the hierarchy is 99.999% accordingly.","PeriodicalId":106240,"journal":{"name":"International Conference on Parallel Processing, 2004. ICPP 2004.","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130122823","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: 7

Mitigating information asymmetries to achieve efficient peer-to-peer queries 减轻信息不对称，实现高效的点对点查询

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327908

Jiang Guo, Baochun Li

引用次数: 0

Energy-efficient scheduling algorithms of object retrieval on indexed parallel broadcast channels 索引并行广播信道中目标检索的节能调度算法

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327953

Bingjun Sun, A. Hurson, J. Hannan

{"title":"Energy-efficient scheduling algorithms of object retrieval on indexed parallel broadcast channels","authors":"Bingjun Sun, A. Hurson, J. Hannan","doi":"10.1109/ICPP.2004.1327953","DOIUrl":"https://doi.org/10.1109/ICPP.2004.1327953","url":null,"abstract":"With the goal of providing \"timely and reliable\" access to information in a mobile computing environment, mobile units and the wireless medium operate under constraints on energy, bandwidth, and connectivity. Among these limitations, power limitation of mobile units is one of the key issues. In a mobile computing environment, broadcasting has proved to be an effective method to distribute public data. Efficient methods for allocating and retrieving objects on parallel indexed broadcast channels have been proposed to manage power consumption and access latency. Employment of parallel channels also brings out the notion of conflicts. To minimize the effect of conflicts on both access latency and power consumption, one has to develop schemes to schedule access to the objects that minimizes the number of passes over the parallel channels. This work extends our past efforts and proposes two new scheduling algorithms that can find the minimum number of passes and inside channel switches. The simulation results show that the proposed scheduling algorithms relative to our previous work have a great impact on energy consumption and access latency. The proposed scheduling algorithms are simulated and results are presented.","PeriodicalId":106240,"journal":{"name":"International Conference on Parallel Processing, 2004. ICPP 2004.","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121844659","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: 13

A reliable connection migration mechanism for synchronous transient communication in mobile codes 移动代码中同步瞬态通信的可靠连接迁移机制

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327952

Xiliang Zhong, Chengzhong Xu

{"title":"A reliable connection migration mechanism for synchronous transient communication in mobile codes","authors":"Xiliang Zhong, Chengzhong Xu","doi":"10.1109/ICPP.2004.1327952","DOIUrl":"https://doi.org/10.1109/ICPP.2004.1327952","url":null,"abstract":"With the increasing popularity of network applications, mobile codes become a promising technology to provide scalable services. Due to their mobile nature, it is a challenge to support synchronous transient communication between mobile objects. This work presents a reliable connection migration mechanism that allows mobile objects in communication to remain connected during their migration. This mechanism supports concurrent migration of both end-points of a connection and guarantees exactly-once delivery of all transmitted data. In addition, a mobile code access control model is integrated to ensure secure connection migration. This work presents the design of the mechanism and a reference implementation, namely NapletSocket, over Java Socket in a mobile agent system. Experimental results show that NapletSocket incurs a moderate cost in connection setup, mainly due to security checking, and marginal overhead for communication over established connections. Furthermore, we investigate the impact of agent mobility on communication performance via simulation. Simulation results show that NapletSocket is efficient for a wide range of migration and communication patterns.","PeriodicalId":106240,"journal":{"name":"International Conference on Parallel Processing, 2004. ICPP 2004.","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127392786","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

An algebra for cross-experiment performance analysis 用于交叉实验性能分析的代数

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327905

Fengguang Song, F. Wolf, N. Bhatia, J. Dongarra, S. Moore

引用次数: 60

Preemption-based avoidance of priority inversion for Java 基于抢占的Java优先级反转避免

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327963

Adam Welc, Antony Lloyd Hosking, S. Jagannathan

{"title":"Preemption-based avoidance of priority inversion for Java","authors":"Adam Welc, Antony Lloyd Hosking, S. Jagannathan","doi":"10.1109/ICPP.2004.1327963","DOIUrl":"https://doi.org/10.1109/ICPP.2004.1327963","url":null,"abstract":"Priority inversion occurs in concurrent programs when low-priority threads hold shared resources needed by some high-priority thread, causing them to block indefinitely. Shared resources are usually guarded by low-level synchronization primitives such as mutual-exclusion locks, semaphores, or monitors. There are two existing solutions to priority inversion. The first, establishing high-level scheduling invariants over synchronization primitives to eliminate priority inversion a priori, is difficult in practice and undecidable in general. Alternatively, run-time avoidance mechanisms such as priority inheritance still force high-priority threads to wait until desired resources are released. We describe a novel compiler and run-time solution to the problem of priority inversion, along with experimental evaluation of its effectiveness. Our approach allows preemption of any thread holding a resource needed by higher-priority threads, forcing it to release its claim on the resource, roll back its execution to the point at which the shared resource was first acquired, and discard any updates made in the interim. The compiler inserts code at synchronization points, permitting rollback of thread execution, and efficient revocation of interim updates. Our design and implementation are realized in the context of IBM's Jikes RVM, a high-quality compiler and runtime system for Java. Our performance results show that throughput of high-priority threads using our scheme can be improved by 30% to 100% when compared with a classical scheduler that does not address priority inversion.","PeriodicalId":106240,"journal":{"name":"International Conference on Parallel Processing, 2004. ICPP 2004.","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124064723","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: 24

The Euler tour technique and parallel rooted spanning tree 欧拉漫游技术与并行根生成树

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327954

Guojing Cong, David A. Bader

{"title":"The Euler tour technique and parallel rooted spanning tree","authors":"Guojing Cong, David A. Bader","doi":"10.1109/ICPP.2004.1327954","DOIUrl":"https://doi.org/10.1109/ICPP.2004.1327954","url":null,"abstract":"Many parallel algorithms for graph problems start with finding a spanning tree and rooting the tree to define some structural relationship on the vertices which can be used by following problem specific computations. The generic procedure is to find an unrooted spanning tree and then root the spanning tree using the Euler tour technique. With a randomized work-time optimal unrooted spanning tree algorithm and work-time optimal list ranking, finding rooted spanning trees can be done work-time optimally on EREW PRAM w.h.p. Yet the Euler tour technique assumes as \"given\" a circular adjacency list, it is not without implications though to construct the circular adjacency list for the spanning tree found on the fly by a spanning tree algorithm. In fact our experiments show that this \"hidden\" step of constructing a circular adjacency list could take as much time as both spanning tree and list ranking combined. We present new efficient algorithms that find rooted spanning trees without using the Euler tour technique and incur little or no overhead over the underlying spanning tree algorithms. We also present two new approaches that construct Euler tours efficiently when the circular adjacency list is not given. One is a deterministic PRAM algorithm and the other is a randomized algorithm in the symmetric multiprocessor (SMP) model. The randomized algorithm takes a novel approach for the problems of constructing the Euler tour and rooting a tree. It computes a rooted spanning tree first, then constructs an Euler tour directly for the tree using depth-first traversal. The tour constructed is cache-friendly with adjacent edges in the tour stored in consecutive locations of an array so that prefix-sum (scan) can be used for tree computations instead of the more expensive list-ranking.","PeriodicalId":106240,"journal":{"name":"International Conference on Parallel Processing, 2004. ICPP 2004.","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124420461","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: 13

Application-aware service differentiation in PAWNs pawn中应用感知的服务差异化

International Conference on Parallel Processing, 2004. ICPP 2004. Pub Date : 2004-08-15 DOI: 10.1109/ICPP.2004.1327950

H. Lufei, S. Sellamuthu, S. Santhosh, Weisong Shi

引用次数: 1