Fraunhofer Institut für Integrierte Publikations-und Informationssysteme IPSI

Filters

22
Reset filters

Settings
Now showing 1 - 10 of 22
  • Publication
    Managing and recovering high data availability in a DHT under chum
    ( 2006)
    Knezevic, P.
    ;
    Wombacher, A.
    ;
    Risse, T.
    An essential issue in peer-to-peer data management is to keep data highly available all the time. A common idea is to replicate data hoping that at least one replica is available when needed. However, due to churns, the number of created replicas could be not sufficient for guaranteeing the intended data availability. If the number of replicas is computed according to the lowest expected peer availability (a classical case), but the expectation were too high, then the peer availability after a churn could be too low, and the system could not be able to recover the requested data availability. The paper is a continuation of previous work [1] and presents an replication protocol that delivers a configured data availability guarantee, and is resistant to, or recovers fast from churns. The protocol is based on a Distributed Hash Table (DHT), measurement of peer online probability in the system, and adjustment of the number of replicas accordingly. The evaluation shows that we are able to maintain or recover the requested data availability during or shortly after stronger or weaker churns, and at the same time the storage overhead is close to the theoretical minimum.
  • Publication
    Managing and recovering high data availability in a DHT under churn
    ( 2006)
    Kneevi, P.
    ;
    Wombacher, A.
    ;
    Risse, T.
    An essential issue in peer-to-peer data management is to keep data highly available all the time. A common idea is to replicate data hoping that at least one replica is available when needed. However, due to churns, the number of created replicas could be not sufficient for guaranteeing the intended data availability. If the number of replicas is computed according to the lowest expected peer availability (a classical case), but the expectation were too high, then the peer availability after a churn could be too low, and the system could not be able to recover the requested data availability. The paper is a continuation of previous work [1] and presents an replication protocol that delivers a configured data availability guarantee, and is resistant to, or recovers fast from churns. The protocol is based on a Distributed Hash Table (DHT), measurement of peer online probability in the system, and adjustment of the number of replicas accordingly. The evaluation shows that we are able to maintain or recover the requested data availability during or shortly after stronger or weaker churns, and at the same time the storage overhead is close to the theoretical minimum.
  • Publication
    Highly available DHTs
    ( 2006)
    Knezevic, P.
    ;
    Wombacher, A.
    ;
    Risse, T.
  • Publication
    Enabling high data availability in a DHT
    ( 2005)
    Kneevi, P.
    ;
    Wombacher, A.
    ;
    Risse, T.
    Many decentralized and peer-to-peer applications require some sort of data management. Besides P2P file-sharing, there are already scenarios (e.g. BRICKS project [3]) that need management of finer-grained objects including updates and, keeping them highly available in very dynamic communities of peers. In order to achieve project goals and fulfill the requirements, a decentralized/ P2P XML storage on top of a DHT (Distributed Hash Table) overlay has been proposed [6]. Unfortunately, DHTs do not provide any guarantees that data will be highly available all the time. A self-managed approach is proposed where availability is stochastically guaranteed by using a replication protocol. The protocol recreates periodically missing replicas dependent on the availability of peers. We are able to minimize generated costs for requested data availability. The protocol is fully decentralized and adapts itself on changes in community maintaining the requested availability. Finally, th e approach is evaluated and compared with replication mechanisms embedded in other decentralized storages.
  • Publication
    Intelligent web service - From web services to Plug&Play service integration
    ( 2005)
    Neuhold, E.
    ;
    Risse, T.
    ;
    Wombacher, A.
    ;
    Niederee, C.
    ;
    Mahleko, B.
    The service oriented architecture and its implementation by Web services have reached a considerable degree of maturity and also a wide adoption in different application domains. This is true for the R&D as well as for the industrial community. Standards for the description, activation, and combination of Web services have been established; UDDI registries are in place for the management of services, and development environments support the software engineer in the creation of Web services. However, the major benefit of service oriented architectures, the loose coupling of services, is still seldom explored in real world settings. The reason is the heterogeneity on different levels within the service oriented architecture. The heterogeneity problems reach from the semantics of service descriptions to compatibility problems between workflows, which have to be connected via service interfaces. In spite of compatible service signatures, workflows might, for example, not be compatible in their semantics. This talk discusses challenges and solutions for a real .Plug&Play. service infrastructure, i.e. a Web service infrastructure, where integration of new Web services becomes as simple and straightforward as plugging a USB stick into your laptop. To achieve this goal various issues have to be addressed: Semantics of services as a foundation for intelligent service mediation and usage Effective, automatic, and intelligent service discovery taking into account application context Dynamic context-aware composition of services into processes The challenges and approaches for a "Plug& Play" service infrastructure are illustrated with a real world example.
  • Publication
    An overview on automatic capacity planning
    ( 2005)
    Risse, T.
    The performance requirement for the transformation of messages within electronic business processes is our motivation to investigate in automatic capacity planning methods. Performance typically means the throughput and response time of a system. Finding a configuration of a distributed system satisfying performance goals is a complex search problem that involves many design parameters, like hardware selection, job distribution and process configuration. Performance models are a powerful tool to analyse potential system configurations, however, their evaluation is expensive, such that only a limited number of possible configurations can be evaluated. In this paper we give an overview of our automatic system design method and discuss the arising problems to achieve the performance during the runtime of the systems. Furthermore we make a discussion on the impact of our strategy on the current trends in distributed systems.
  • Publication
    Understanding and tailoring your scientific information environment: A context-oriented view on e-science support
    ( 2005)
    Niederée, C.
    ;
    Stewart, A.
    ;
    Muscogiuri, C.
    ;
    Hemmje, M.
    ;
    Risse, T.
  • Publication
    Supporting information access in next generation digital library architectures
    ( 2005)
    Frommholz, I.
    ;
    Knezevic, P.
    ;
    Mehta, B.
    ;
    Niederée, C.
    ;
    Risse, T.
    ;
    Thiel, U.
    Current developments on Service-oriented Architectures, Peer-to-Peer and Grid computing promise more open and flexible architectures for digital libraries. They will open the Digital Library (DL) technology to a wider clientele, allow faster adaptability and enable the usage of federative models on content and service provision. These technologies raise new challenges for the realization of DL functionalities, which are rooted in the increased heterogeneity of content, services and metadata, in the higher degree of distribution and dynamics, as well as in the omission of a central control instance. This paper discusses these opportunities and challenges for three central types of DL functionality revolving around information access: metadata management, retrieval functionality, and personalization services.
  • Publication
    A self-organizing data store for large scale distributed infrastructures
    ( 2005)
    Risse, T.
    ;
    Kneevi, P.
    The introduction of Service Oriented Architectures enables the transition from integrated, centrally controlled systems to federated and dynamic configurable systems. Scalability and reliability are key requirements for such infrastructures. Due to the size and dynamic only decentralized and self-organizing systems are able to accomplish the previously mentioned requirements. In this paper we describe our approach to tackle the problem within the BRICKS project. Here we use a completely decentralized XML database to store several types of metadata. We will present the database architecture and the way it is used in the BRICKS architecture, exemplary for service discovery. Furthermore will we give an overview of current research activities and preliminary results.