Search

Environmental pollution control relies heavily on human expert judgment sup- ported by historical data and scientific models. Telemonitoring, by networks of heterogeneous sensor arrays, provides the opportunity for data mining models to be constructed from the historical data to supplement human expertise. This paper reports some progress made in the TELEMAC project by data mining. TELEMAC is concerned with enhancing the efficacy of anaerobic digestion in potentially unstable digesters. In the laboratory using full instrumentation it is possible to derive a good description of the digester state. With data mining it is possible to identify some constraints on sensor choice. This paper examines this data mining work from the perspective of a three layer Grid architecture to see what implications and requirements arise that could benefit the exercise of expert judgment. After placing the specific TELEMAC situation in a generic Grids context, we present a classification approach to attributes for metadata and indicate some examples of model resource discovery. [ABSTRACT FROM AUTHOR]

In this paper we address the problem of mining frequent closed itemsets in a highly distributed setting like a Grid. The extraction of frequent (closed) item-sets is a very expensive phase needed to extract from a transactional database a reduced set of meaningful association rules. We figure out an environment where different datasets are stored in different sites. We assume that, due to the huge size of datasets and privacy concerns, dataset partitions cannot be moved to a centralized site where to materialize the whole dataset and perform the mining task. Thus it becomes mandatory to perform separate mining at each site, and then merge local results for deriving global knowledge. This paper shows how frequent closed itemsets, mined independently at each site, can be merged in order to derive globally frequent closed itemsets. Unfortunately, such merging might produce a superset of all the frequent closed itemsets, while the associated supports could be smaller than the exact ones because some globally frequent closed itemsets might be not locally frequent in some partitions. To avoid an expensive post-processing phase, needed to compute exact global results, we use a method to approximate the supports of closed itemsets. The approximation is only needed for those globally (closed) frequent itemsets which are locally infrequent on some dataset partitions, and thus are not returned at all from the corresponding sites. [ABSTRACT FROM AUTHOR]

We study the problem of resource discovery in the Semantic Grid. We show how to solve this problem by utilizing Atlas, a P2P system for the distributed storage and retrieval of RDF(S) data. Atlas is currently under development in project OntoGrid funded by FP6. Atlas is built on top of the distributed hash table Bamboo and supports pull and push querying scenarios. It inherits all the nice features of Bamboo (openness, scalability, fault-tolerance, resistance to high churn rates) and extends Bamboo’s protocols for storing and querying RDF(S) data. Atlas is being used currently to realize the metadata service of S-OGSA in a fully distributed and scalable way. In this paper, we concentrate on the main features of Atlas and demonstrate its use for Semantic Grid resource discovery in an OntoGrid use case scenario. [ABSTRACT FROM AUTHOR]

Computational Grids can be effectively used as an infrastructure for distributed data mining and knowledge discovery in large data sets. To utilize Grids for high-performance knowledge discovery, software tools and mechanisms are needed. To this purpose we designed a system called Knowledge Grid and we are implementing its services as WSRF-compliant Grid Services. This chapter describes the composition of distributed knowledge discovery services, according to the service oriented architecture model, by using the Knowledge Grid environment. We discuss Grid Services for searching Grid resources, composing software and data elements, and executing the resulting data mining application on the Knowledge Grid. The chapter focuses in particular on the application modeling. Applications are designed using a UML model, which is translated into a BPEL representation, in turn processed by the Knowledge Grid services for its execution. [ABSTRACT FROM AUTHOR]

Co-ordinated usage of resources in a Grid environment is a challenging task impeded by the nature of resource usage and provision: Resources reside in different geographic locations, are managed by different organisations, and are by no means accessible via standardised interfaces, protocols or commands. These prerequisites have to be taken into account in order to provide solutions in the area of Grid scheduling and resource management. In this document we propose the employment of a semantic model for Grid scheduling. The Grid Scheduling Ontology describes the capabilities and state of a scheduler, providing a machine-processable and interoperable model for the integration of local schedulers into Grid resource management. Along with the model we present a meta-scheduling architecture based on the VIOLA Meta-Scheduling Service that uses ontology modelling and reasoning capabilities of OWL to provide semantic support for meta-scheduling in Grids. [ABSTRACT FROM AUTHOR]

Ontologies are the backbone of the Semantic Web. Current grid architectures do not consider their usage, and there are no protocols nor standards in the Grid community for dealing with them. Therefore, the provision of appropriate means for accessing, querying and using ontologies effectively is a key factor if we want to enrich the current grid with semantic technologies and to support progress towards the next generation Grid, that is, the Semantic Grid. [ABSTRACT FROM AUTHOR]

A general model for the representation of scientific study metadata does not exist. The e-Science enablement of the data holdings of CCLRC requires such a model to allow access to the data resources of the facilities in a uniform way. By proposing a model and an implementation, the adoption of such a system would aid interoperability of scientific information systems in the organisation and form a specification of the type and categories of metadata that studies should capture about their investigations and the data they produce inside and outside of CCLRC. This allows further exploitation of scientific Studies and associated datasets, ease citation, facilitate collaboration and allow the easy integration of pre-Grid metadata into a common Grid/e-Science enabled scientific information platform. In this paper, we describe a science metadata model developed at CCLRC, with its motivation, overall design, usage and future development. [ABSTRACT FROM AUTHOR]

The Semantic Grid reference architecture, S-OGSA, includes semantic provisioning services that are able to produce semantic annotations of Grid resources, and semantically aware Grid services that are able to exploit those annotations in various ways. In this paper we describe the dynamic aspects of S-OGSA by presenting the typical patterns of interaction among these services. A use case for a Grid meta-scheduling service is used to illustrate how the patterns are applied in practice. [ABSTRACT FROM AUTHOR]

Nowadays grid applications process large volumes of data. This creates the need for an effective data-management solutions. For the ClusteriX project the CDMS (ClusteriX Data Management System) is being developed. Analysis of user requirements and existing implementations of a Data Management System have been the foundations for its creation. Special attention has been paid to make the system user-friendly and efficient. Taking into account grid specific networking conditions, for example different bandwidth, current load and network technologies, between geographically distant sites, CDMS tries to optimize data throughput via replication and replica selection techniques. Another key feature to be considered during grid service implementation is fault-tolerance. In the CDMS modular design and distributed operation model assures single point of failure elimination. In particular multiple instances of Data Broker are running simultaneously and their coherence is assured by a synchronization subsystem. [ABSTRACT FROM AUTHOR]

The quality of virtualization mechanisms provided by a storage system affects storage management complexity and storage efficiency, both of which are important problems of modern storage systems. We argue that current storage systems provide limited flexibility and extensibility in virtualizing, managing and accessing storage. In this work we address this problem by proposing Violin, a virtualization framework that allows easy extensions of block-level storage stacks. Violin allows (i) developers to provide new virtualization functions and (ii) storage administrators to combine these functions in storage hierarchies with rich semantics. Violin makes it easy to develop new virtualization functions by providing support for (i) hierarchy awareness and arbitrary mapping of blocks between virtual devices, (ii) an easily extensible I/O command set, (iii) explicit control over both the request and completion path of I/O requests, and (iv) persistent metadata management. In this paper we present Violin’s architecture and we show how simple Violin modules can be combined in more complex hierarchies. Finally, we demonstrate hierarchies with advanced virtualization functionality that is difficult to implement with monolithic drivers. [ABSTRACT FROM AUTHOR]

Scalable storage systems are expected to scale to large numbers of physical storage devices and to service diverse applications without incuring high management costs. New storage virtualization architectures and techniques that are currently being proposed, aim at addressing these needs by providing the ability to configure storage systems to meet resource constraints and application requirements. However, this flexibility leads to a large number of options when configuring storage systems either statically or dynamically. In this work we examine how this process can be automated. We present Conductor, a rule-based production system that is able to evaluate alternatives and minimize system cost, based on certain criteria. Conductor starts from a set of system resources and a set of application requirements and proposes specific system configurations that meet application requirements while minimizing resource costs. It captures human expertise in the form of rules to generate and evaluate configuration alternatives. In this work we focus on static configuration issues and examine various approaches for reducing complexity within a large configuration space. Our techniques manage to satisfy practical time and resource constraints. [ABSTRACT FROM AUTHOR]

Fault-tolerance has long been a feature of database systems, with transactions supporting the structuring of applications so as to ensure continuation of updating applications in spite of machine failures. For read-only queries the perceived wisdom has been that support for fault-tolerance is too expensive to be worthwhile. Distributed query processing (DQP) is coming to be seen as a promising way of implementing applications that combine structured data and analysis operations in dynamic distributed settings such as computational grids. Accordingly, a number of protocols have been described that support tolerance to failure of intermediate machines, so as to permit continuation from surviving intermediate state. However, a distributed query can have a non-trivial mapping onto hardware resources. Because of this it is often possible to choose between a number of possible recovery strategies in the event of a failure. The work described here makes an initial investigation in this area in the context of an example query expressed over distributed resources in a Grid and shows that it can be worthwhile to make this choice between recovery alternatives dynamically, at the point a failure is detected rather than statically beforehand. [ABSTRACT FROM AUTHOR]

The BioDA project is investigating how Bioinformatics Grids, a data and compute intensive domain, could benefit from using a standard framework, such as OGSA-DAI, to manage access and integration of distributed heterogeneous data resources. In this paper, we outline the common data access and integration requirements from the bioinformatics community. We then highlight some specific issues encountered while designing an OGSA-DAI exemplar application for Biodiversity World, a Biodiversity Grid that specialises in extracting knowledge from correlating a plethora of distributed heterogeneous data sources accesible via the Web in the study of biodiversity patterns. [ABSTRACT FROM AUTHOR]

To date there have been several efforts with a view to developing services that support and enable data integration on the Grid; however there is a lack of a comprehensive solution to this issue. This paper summarises the work thus far on the XMAP data integration framework and query reformulation algorithm and on middleware with regard to Grid query processing services, namely OGSA-DQP. Furthermore, it presents an architecture for data integration-enabled query processing on the Grid, which combines the two aforementioned pieces of work and provides an extended set of e-Services. These services allow users to submit queries over a single database and receive the results from multiple databases that are semantically correlated with the former one. The paper focuses on the service choreography involved by elaborating on the interactions between the services, and discusses the extensions to OGSA-DQP that are required in order to make the services interoperable. [ABSTRACT FROM AUTHOR]

The grid provides a vision in which resources, including storage and data, can be shared across organisational boundaries. The original emphasis of grid computing lay in the sharing of computational resources but technological and scientific advances have led to an ongoing data explosion in many fields. However, data is stored in many different storage systems and data formats, with different schema, access rights, metadata attributes, and ontologies all of which are obstacles to the access, integration and management of this information. In this chapter we examine some of the ways in which these differences can be addressed by grid technology to enable the meaningful sharing of data. In particular, we present an overview of the OGSA-DAI (Open Grid Service Architecture - Data Access and Integration) software, which provides a uniform, extensible framework for accessing structured and semi-structured data and provide some examples of its use in other projects. The open-source OGSA-DAI software is freely available from http://www.ogsadai.org.uk. [ABSTRACT FROM AUTHOR]