2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)最新文献

{"title":"Multiple Query Optimization for Data Analysis Applications on Clusters of SMPs","authors":"H. Andrade, T. Kurç, A. Sussman, J. Saltz","doi":"10.1109/CCGRID.2002.1017123","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017123","url":null,"abstract":"This paper is concerned with the efficient execution of multiple query workloads on a cluster of SMPs. We target applications that access and manipulate large scientific datasets. Queries in these applications involve user-defined processing operations and distributed data structures to hold intermediate and final results. Our goal is to implement system components to leverage previously computed query results and to effectively utilize processing power and aggregated I/O bandwidth on SMP nodes so that both single queries and multi-query batches can be efficiently executed.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114640895","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}

{"title":"Monitoring and Controlling Distributed Applications with Relocatable Objects","authors":"S. Gilmore, Marco A. Palomino","doi":"10.1109/CCGRID.2002.1017158","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017158","url":null,"abstract":"The Java programming language and its environment are highly regarded as suitable choices for the development and deployment of distributed applications. Portability, an object oriented approach, support for multi-threading programming and a special set of libraries for remote procedure calls, are some of the features that make Java an excellent choice for the implementation of distributed applications. A considerable number of projects on Java distributed systems have already been completed. Here we can list Charlotte, SuperWeb, Javelin, Javelin++, Java/DSM, Voyager and Babylon. Based on Java, the Babylon package [1] was designed to integrate into a single comprehensive system both new and previously researched ideas in Java distributed computing. It retains parts of the underlying mechanisms used in Ajents [2], a previous project from which Babylon took part of its source code. Babylon also has some innovative additions that are not currently supported by Java, such as remote object creation, remote class loading, asynchronous remote method invocation and object migration. From the point of view of application designers, Babylon can be deployed as a layer of middleware which facilitates the use of clusters of computational resources allowing selective migration of code-containing objects from one computational environment to another. From the point of view of programmers, Babylon can be seen as a collection of Java class libraries that provides the necessary software infrastructure to support straightforward access to information that is distributed, within organisations and all around the world. It is relevant to note that all of the Babylon class libraries are written in pure Java. Among other things, this means that Babylon can be executed on any standard Java Virtual Machine (JVM), and it is potentially open to a vast audience of programmers. Besides, the Babylon package contains some useful features that are not currently supported by Java. Consequently, it reduces the difficulty of writing distributed programs. Babylon is implemented using Java RMI. It has been shown, however, that Java RMI is very slow for high performance computing [4]. Most of the researchers in the field agree that it is inappropriate for interactive applications [3]. Given that the key features of Babylon are principally based on Java RMI, all the drawbacks and limitations of Java RMI are inherited by Babylon. The motivation behind this work is the submission of a contribution that eventually helps Babylon to become a better tool in the field of Java distributed object applications. To this end, we have developed graphical components to provide monitoring and management capabilities for a Babylon application. We have also made a number of technical extensions to the existing Babylon system, including upgrading it to the Java 2 security model, and we have adapted it to use a better implementation of RMI, because we believe that its dependency on Sun’s RMI represents a","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116041792","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}

{"title":"Nomadic Migration: Fault Tolerance in a Disruptive Grid Environment","authors":"Gerd Lanfermann, Gabrielle Allen, T. Radke, E. Seidel","doi":"10.1109/CCGRID.2002.1017143","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017143","url":null,"abstract":"Nomadic Migration describes a technology, which provides an application with the ability to seek out and exploit remote computing resources by migrating tasks from site to site, dynamically adapting the application to a changing Grid environment. By automating the detection and usage of free resources in a global Grid, we achieve a significantly faster throughput than by manually interfacing with these resources. In this Paper we discuss the Peer-To-Peer strategy as an approach to provide a fault tolerant service infrastructure, required for a stable Grid Migration Service in an intrinsically disruptive Grid environment. The migration technology presented here is e.g. used with large-scale, Cactus based HPC simulations.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122074036","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}

{"title":"Handling Multimedia Objects in Peer-to-Peer Networks","authors":"V. Kalogeraki, A. Delis, D. Gunopulos","doi":"10.1109/CCGRID.2002.1017184","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017184","url":null,"abstract":"Video-on-demand systems and services [4, 2] are predominantly offered over dedicated private networks with the help of large servers [3, 13]. Such systems are restricted by the number of concurrent accesses allowed as well as load balancing issues that ensue when the demand for video objects is skewed [1, 13]. The widespread usage of broadband networks in connection with the low-cost commodity hardware may offer alternative avenues for the delivery of video streams and other multimedia objects. This goal can be achieved with the help of multiple, independent, and inexpensive computing nodes that function using a peer-to-peer P2P protocol [9, 8, 7]. P2P services have so far concentrated in the exchange/sharing of “small” objects including mp3, images, and audio files as well as CPU cycles and memory space [5, 11]. Discovery of objects solely based on the existing message-based schemes may prove rather ineffective [10]. Centralized indexing methods used in early P2P systems are very restrictive [12]. As groups of sites tend to form sizable node clusters [10], the large total number of dispatched messages makes inefficient use of computing resources. It is our position that furnishing video services on a P2P network not only can offer viable alternatives to the mostly proprietary architectures used today for the delivery of video services, but also it can be done in a reliable and scalable manner. We build upon the approach of ad-hoc P2P networks of resources and propose a new architecture that can support storage and retrieval of movies and/or video-clips. Our proposed configuration exploits the availability of high-performance links to networks, the usage of exclusive and partial indexing in peers, making nodes “aware” of the content in their own vicinity, replication of objects and caching of popular items, as well as full connectivity among servers whenever feasible. The architecture uses efficient indexing mechanisms for the retrieval of the multimedia objects, guarantees continuous operation in light of server failures and allows for the transparent population of new servers as well as the evolution of the furnished services and/or network resources with minimal disruption to the users. One key provision to realize such a P2P infrastructure is that the core peers (or servers) are expected to be linked via a low-latency and high-bandwidth network capable of effective handling and delivery of voluminous multimedia data. It is also anticipated that end-users should have sufficient connections to object servers. For acceptable quality MPEG-compressed video streams, one would approximately need 30 frames/sec and around 1.5Mbit/sec network connection to the source. This is well within reach as more individuals are connected through specialized modems (cable, DSL, etc.) or T1-level network connections [14, 6]. The system consists of a number of servers that constitute the basis for managing the storage and retrieval of video objects. Each peer functi","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"2022 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126281706","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}

{"title":"Study of Load Balancing Issues Based on Intra-Movie Skewness for Parallel Video Servers","authors":"Song Wu, Hai Jin","doi":"10.1109/CCGRID.2002.1017137","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017137","url":null,"abstract":"In order to avoid the load imbalance problem caused by video popularity, parallel video servers divide video objects into small segments and put them across multiple server nodes. However, due to users’ various viewing time, access numbers of movie segments are quite different. Some segments are more popular than others. This instance is called intra-movie skewness, which may leads to load imbalance among server nodes in parallel video servers. In this paper we analyze and model the intra-movie skewness. Then, a novel data placement strategy, SPS (Symmetrical Pair Scheme), is proposed. It is proved that SPS prevents the impact of intra-movie skewness and has better load balancing performance than traditional round robin data placement, especially in a large-scale parallel video server.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114137975","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}

{"title":"On the Migration of the Scientific Code Dyana from SMPs to Clusters of PCs and on to the Grid","authors":"M. Taufer, T. Stricker, Gerard Roos, P. Güntert","doi":"10.1109/CCGRID.2002.1017116","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017116","url":null,"abstract":"Dyana is a molecular biology code used in the study of infectious prion proteins. Like many other scientific codes, Dyana was migrated successfully from vector supercomputers to some more cost-effective cluster of commodity PCs. A further migration to a widely distributed grid computing platform looks very tempting because many of these platforms promise the use of nearly free compute-cycles on the Internet. Not all codes are equally suited for all platforms. Even embarrassingly parallel codes might require a significant re-engineering effort for a migration from one platform to another. A better understanding of the performance characteristics of a code is required before a migration is attempted. To address this problem, we present a systematic method to study the viability of a code migration from one platform to another, before it is actually undertaken. We construct an analytic performance model of the application. We use the previous migration from SMPs to commodity clusters of PCs to validate and calibrate the model. Finally, we extrapolate the performance of Dyana widely distributed computing on the grid and we suggest optimizations in the process of migration. Our general model predicts that Dyana can efficiently use up to 42000 processors with its current workload and is therefore well suited for grid computing on the Internet.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132689892","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}

{"title":"Maintaining Connectivity in a Scalable and Robust Distributed Environment","authors":"Márk Jelasity, M. Preuss, M. Steen, B. Paechter","doi":"10.1109/CCGRID.2002.1017166","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017166","url":null,"abstract":"This paper describes a novel peer-to-peer (P2P) environment for running distributed Java applications on the Internet. The possible application areas include simple load balancing, parallel evolutionary computation, agent-based simulation and artificial life. Our environment is based on cutting-edge P2P technology. We introduce and analyze the concept of long term memory which provides protection against partitioning of the network. We demonstrate the potentials of our approach by analyzing a simple distributed application. We present theoretical and empirical evidence that our approach is scalable, effective and robust.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132913210","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}

{"title":"A Report from the U.S. National Science Foundation Blue Ribbon Panel on Cyberinfrastructure","authors":"D. Atkins","doi":"10.1109/CCGRID.2002.1017106","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017106","url":null,"abstract":"This talk will review the mission, activities, and recommendations of the “Blue Ribbon Panel on Cyberinfrastructure” recently appointed by the leadership on the U.S. National Science Foundation (NSF). The NSF invests in “people, ideas, and tools” and in particular is a major investor in basic research to produce communication and information technology (ICT) as well as its use in supporting basic research and education in most all areas of science and engineering. The NSF through its Directorate for Computer and Information Science and Engineering (CISE) has provided substantial funding for high-end computing resources, initially by awards to five supercomputer centers and later through $70 M per year investments in two partnership alliances for advanced computation infrastructures centered at the University of Illinois and the University of California, San Diego. It has also invested in an array of complementary R&D initiatives in networking, middleware, digital libraries, collaboratories, computational and visualization science, and distributed terascale grid environments.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131872838","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}

{"title":"Towards Compliant Distributed Shared Memory","authors":"K. Falkner, Henry Detmold, David S. Munro, Travis Olds","doi":"10.1109/CCGRID.2002.1017151","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017151","url":null,"abstract":"There exists a wide spectrum of coherency models for use in distributed shared memory (DSM) systems. The choice of model for an application should ideally be based on the application's data access patterns and phase changes. However, in current systems, most, if not all of the parameters of the coherency model are fixed in the underlying DSM system. This forces the application either to structure its computations to suit the underlying model or to endure an inefficient coherency model. This paper introduces a unique approach to the provision of DSM based on the idea of compliance. Compliance allows an application to specify how the system should most effectively operate through a separation between mechanism, provided by the underlying system, and policy, pro-vided by the application. This is in direct contrast with the traditional view that an application must mold itself to the hard-wired choices that its operating platform has made. The contribution of this work is the definition and implementation of an architecture for compliant distributed coherency management. The efficacy of this architecture is illustrated through a worked example.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115960446","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}

{"title":"Experiments of Network Throughput Measurement and Forecasting Using the Network Weather","authors":"P. Primet, R. Harakaly, Franck Bonnassieux","doi":"10.1109/CCGRID.2002.1017171","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017171","url":null,"abstract":"A tool able to predict the behavior of the network in a Global computing environment could help the users to obtain better performance. In this paper we study the accuracy of the Network Weather Service (NWS) networking features and specifically the TCP throughput measurement and prediction. We studied the NWS forecast stability and confidence level on different types of networking connectivity. To examine how the predicted values can be applied to \"bulk\" transfer, the prediction was compared with results of measurements achieved with the Iperf network monitoring tool. Iperf results are more accurate than the NWS ones but the tool is much more intrusive. We propose an improvement to use its predicting feature in an internet based global computing environment.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129876545","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}