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

{"title":"Developing Distributed Data Mining Implementations for a Grid Environment","authors":"Wei Du, G. Agrawal","doi":"10.1109/CCGRID.2002.1017185","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017185","url":null,"abstract":"1 Introduction Distributed Data Mining (DDM) is the process of analyzing geographically dispersed large datasets for extracting novel and interesting patterns or models [5]. This paper reports our preliminary work in a project that takes a new approach towards distributed data mining. We believe that general grid application development tools and techniques [4] should be used for developing mining implementations that analyze distributed datasets. We use Dat-aCutter, which is a general grid middleware for data intensive problems and is based upon filter-stream programming model [1]. In this paper, we present a case study of developing k-nearest neighbor search using this framework. Our experiments show that filter granularity selection and filter placement are important issues in implementing data mining algorithms with DataCutter. We are also developing language and compiler support to offer a high-level interface for DataCutter. A data parallel dialect of Java is proposed which is suitable for expressing a variety of data mining tasks. This paper also describes our initial ideas for translating a program written in data parallel Java to a filter-stream program that could be run on DataCutter. 2 A Case Study To validate our belief that distributed data mining implementations can be conveniently developed using DataCut-ter, we have implemented a distributed version of k-nearest neighbor classifier. The problem is quite simple: given a 3-dimension range R = ´Ü ½ Ý ½ Þ ½ µ ´Ü ¾ Ý ¾ Þ ¾ µ , and a point w = (a,b,c), where x ½ ,y ½ ,z ½ are the coordinates of the upper left corner of R and x ¾ ,y ¾ ,z ¾ are the coordinates of the lower right corner of it, we want to find the nearest k neighbors of w within range R, where k is a parameter given to the algorithm. The basic idea for finding k-nearest neighbors in a sequential environment is as follows. For each point X = (x,y,z) in the file, if X is in the range R, we compute the distance from X to w by dist =","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"117 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":"114483300","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":"Optimistic Synchronization and Transactional Consistency","authors":"M. Wende, M. Schöttner, R. Göckelmann, T. Bindhammer, P. Schulthess","doi":"10.1109/CCGRID.2002.1017155","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017155","url":null,"abstract":"Distributed Shared Memory (DSM) is an interesting alternative to build distributed object system. Explicit message passing and remote invocation is replaced by the simple and unified DSM abstraction. The recurrent performance drawbacks of DSM Systems are often caused by expensive distributed locking mechanisms. In response to this our multipurpose Plurix Operating System (OS) implements a Transaction based DSM. Memory consistency is maintained by atomic transactions and optimistic synchronization mechanisms which have been used in database technology in the past. Such a transaction based DSM with optimistic synchronization guarantee a sequential consistent view on the shared data to every mode in the cluster.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"5 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":"114007524","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 Trust-Aware Resource Management in Grid Computing Systems","authors":"Farag Azzedin, Muthucumaru Maheswaran","doi":"10.1109/CCGRID.2002.1017189","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017189","url":null,"abstract":"Resource management is a central part of a Grid computing system. In a large-scale wide-area system such as the Grid, security is a prime concern. One approach is to be conservative and implement techniques such as sandboxing, encryption, and other access control mechanisms on all elements of the Grid. However, the overhead caused by such a design may negate the advantages of Grid computing. This study examines the integration of the notion of \"trust\" into resource management such that the allocation process is aware of the security implications. We present a formal definition of trust and discuss a model for incorporating trust into Grid systems. As an example application of the ideas proposed, a resource management algorithm that incorporates trust is presented. The performance of the algorithm is examined via simulations.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"82 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":"123048416","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":"End-Host Multicast Communication Using Switch-Trees Protocols","authors":"David A. Helder, S. Jamin","doi":"10.1109/CCGRID.2002.1017172","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017172","url":null,"abstract":"Switch-trees are peer-to-peer algorithms for building and improving end-host multicast trees. Nodes switch parents to reduce tree cost or lower source-member latency. A node switches parents by disconnecting from its parent and reconnecting to a new parent. If the new parent is well chosen, the performance of the tree is improved overall. We look at the performance of switch-trees using the following metrics: cost, latency, link stress and number of switches. Simulations show switch-tree algorithms can build trees of hundreds of nodes at less than twice the optimal cost. In addition, we describe our implementation of a switch-tree protocol. Experiments show that our protocol builds low-cost trees in practice.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"74 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":"129301512","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":"Agents for the Grid: A Comparison with Web Services (Part I: Transport Layer)","authors":"L. Moreau","doi":"10.1109/CCGRID.2002.1017130","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017130","url":null,"abstract":"The notion of agent has of late become popular in the Grid community, as exemplified by several workshops on the use of agents in the Grid. What are agents for the Grid? What is the difference between agents and Web-services? These are questions that we address by describing a port of the SoFAR agent framework to Web services in the context of a bioinformatics Grid. In this first paper, we focus our discussion solely on issues at the transport layer. Through an agent communication language (ACL) and an abstract communication model, we have been able to define a generic API to communications, and are able to support multiple protocols, including the XML protocol, the transport mechanism of Web services. This approach facilitates the development of applications, makes our environment future-proof, and promotes the open-ness of our Grid architecture to third-party developers.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"246 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":"116386446","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":"Exploiting Web Technologies for Grid Architecture","authors":"Gang Xue, M. Fairman, S. Cox","doi":"10.1109/CCGRID.2002.1017139","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017139","url":null,"abstract":"Open-standard Web technologies, such as XML and Web Services, offer new opportunities for the construction of Grid infrastructures that are flexible and extensible. As the Grid community focuses on the development of open grid computing protocols, it is important to consider how existing (and emerging) open standard technologies could be exploited and deployed effectively on the Grid. In this paper we introduce our work on applying the Web technologies for the construction of a Grid component that wraps the Condor distributed computing system and its application in a Grid scenario. Furthermore we propose a model for building Web Services on the Grid","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":"126511668","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 Novel Architecture for Grid Information Systems","authors":"Z. Balaton, G. Gombás, Z. Németh","doi":"10.1109/CCGRID.2002.1017140","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017140","url":null,"abstract":"Grids facilitate large-scale distributed resource sharing.In such an environment a priori knowledge is not availabledue to the diversity of resources and their dynamics. In-formation in the grid ranges from static to highly dynamic.We focus on information required for resource brokering,i.e. how appropriate resources for applications can be foundand what requirements it poses for an information system.One of the representative grid information systems is theGlobus Metacomputing Directory Service (MDS) currentlyin its second incarnation called MDS-2 [1]. It performs bet-ter than its predecessor with respect to scalability and per-formance, yet it still may fail to satisfy every needs of aresource broker. In MDS-2 Grid Information Index Ser-vice (GIIS) servers (specialised indexes of the available re-sources) are used to provide a central repository for clientsto facilitate searches. They collect, cache and manage in-formation from resources belonging to a so called virtualorganisation. It works well in small virtual organisationsbut it is not scalable for thousands of resources.One of the reasons is that MDS-2 aims to provide a uni-form information and monitoring system, i.e. handles staticand dynamic data in the same way. Since frequently chang-ing data becomes stale quickly, the tree of the GIIS serversforminga cache chain cannot be tall. Also as a consequenceof the pull data delivery model used in MDS-2 the amountofdata movedis proportionalto the numberof queries. Thislimits scalability and efficiency.Furthermore, MDS-2 uses the LDAPv3 [4] protocolwhere information is organised as a hierarchical tree calledDirectory Information Tree (DIT) defined by the distin-guished names (DN) of the entries. Although the LDAPquery language permits searches based on any properties ofthe entries, a query that does not match the hierarchy of thedistributed LDAP database can be very inefficient. This isbecause the DIT is distributed along the hierarchy definedby the DNs and one can only restrict the set of servers tobe searched in terms of this hierarchy (with the base andscope parameters in the query). Thus, most queries whichdo not match the hierarchy predefined by the DNs of en-tries result in querying every server storing the distributedLDAP database. Therefore, the distributed LDAP databasecan only answer certain searches efficiently that match thepredefined hierarchy.The solution proposed in this paper tries to tackle withthese problems. Key points are introduced in the following.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"49 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":"122802003","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":"Grids Deployment at the Crossroads - An Update on the EU-Funded Research Efforts","authors":"Kyriakos Baxevanidis","doi":"10.1109/CCGRID.2002.1017107","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017107","url":null,"abstract":"Grid technologies are establishing today the basis of a new class of infrastructure that will enable a new generation of (revolutionary) approaches to scientific research. At the same time, Grid prospects look more than ever before so close to evolving commercial needs. In this picture, current research initiatives clearly demonstrate a trend to further develop and evolve Grid computational concepts in an effort to link the Grid to a much broader range of problems, application areas, and other technological concepts. It is obvious that the impact of Grid technology on the broader e-Economy domain needs further understanding. That may lead to important decisions on a national or international level regarding relevant major infrastructure upgrades.","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":"123174620","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":"MS-I/O: A Distributed Multi-Storage I/O System","authors":"X. Shen, A. Choudhary","doi":"10.1109/CCGRID.2002.1017124","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017124","url":null,"abstract":"More and more parallel applications are running in a distributed environment to take advantage of easily available and inexpensive commodity resources. For data intensive applications, employing multiple distributed storage resources has many advantages. In this paper, we present a Multi-Storage I/O System (MS-I/O) that can not only effectively manage various distributed storage resources in the system, but also provide novel high performance storage access schemes. MS-I/O employs many state-of-the-art I/O optimizations such as collective I/O, asynchronous I/O etc. and a number of new techniques such as data location, data replication, subfile, superfile and data access history. In addition, many MS-I/O optimization schemes can work simultaneously within a single data access session, greatly improving the performance. Although I/O optimization techniques can help improve performance, it also complicates I/O system. In addition, most optimization techniques have their limitations. Therefore, selecting accurate optimization policies requires ex-pert knowledge which is not suitable for end users who may have little knowledge of I/O techniques. So the task of I/O optimization decision should be left to the I/O system itself, that is, automatic from user's point of view. We present a User Access Pattern data structure which is associated With each dataset that can help MS-I/O easily make accurate I/O optimization decisions.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"18 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":"130460497","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":"Efficient Utilization of Memory Mapped NICs onto Clusters using Pipelined Schedules","authors":"A. Sotiropoulos, G. Tsoukalas, N. Koziris","doi":"10.1109/CCGRID.2002.1017132","DOIUrl":"https://doi.org/10.1109/CCGRID.2002.1017132","url":null,"abstract":"This paper describes the performance benefits attained using enhanced network interfaces to achieve low latency communication. We make use of DMA communication mode, to send data to other nodes, while the CPU performs useful calculations. Zero-copy communication is achieved through pinned-down physical memory regions, provided by NIC's driver modules. Our testbed concerns the parallel execution of tiled nested loops onto a Linux PC cluster with PCI-SCI NICs (Dolphin D330). Tiles are essentially exchanging data and should also have large Computational grain, so that their parallel execution becomes beneficial. We schedule tiles much more efficiently by exploiting the inherent overlapping between communication and computation phases among successive, atomic tile executions. The applied nonblocking schedule resembles a pipelined data-path where computation phases are overlapped with communication ones, instead of being interleaved with them. Experimental evaluation illustrates that when using enhanced communication features such as DMA transfers, memory-mapped interfaces and zero-copy mechanisms, overall performance is considerably improved compared to using conventional, CPU and kernel bounded, communication primitives.","PeriodicalId":160584,"journal":{"name":"2nd IEEE/ACM International Symposium on Cluster Computing and the Grid (CCGRID'02)","volume":"29 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":"115036865","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}