Self-tuning management of update-intensive multidimensional data in clusters of workstations

Kriakov, Vassil; Kollios, George; Delis, Alex

doi:10.1007/s00778-008-0121-2

Cited by 2 publications

(1 citation statement)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We believe that DBDs are orthogonal to all of these phases, meaning that they could be used to describe them as a whole or separately. Our initial focus is not on schema integration, nor data cleansing, nor self-tuning [36] DBMS -rather, DBDs offer a means for DBMS self-description. In this sense, they can be used by mediators in a federation.…”

Section: Related Workmentioning

confidence: 99%

Organization is sharing

Senra¹

View full text Add to dashboard Cite

Information sharing has always been a key issue in any kind of joint effort. Paradoxically, with the data deluge, the more information available, the harder it is to design and implement solutions that effectively foster such sharing. This thesis analyzes distinct aspects of sharing -from eScience-related environments to personal information. As a result of this analysis, it provides answers to some of the problems encountered, along three axes.The first, SciFrame, is a specific framework that describes systems or processes involving scientific digital data manipulation, serving as a descriptive pattern to help system comparison. The adoption of SciFrame to describe distinct scientific virtual environments allows identifying commonalities and points for interoperation.The second axe contribution addresses the specific problem of communication between arbitrary systems and services provided by distinct database platforms, via the use of the so-called database descriptors or DBDs. These descriptors contribute to provide independence between applications and the services, thereby enhancing sharing across applications and databases.The third contribution, Organographs, provides means to deal with multifaceted information organization. It addresses problems of sharing personal information by means of exploiting the way we organize such information. Here, rather than trying to provide means to share the information itself, the unit of sharing is the organization of the information. By designing and sharing organographs, distinct groups provide each other dynamic, reconfigurable views of how information is organized, thereby promoting interoperability and reuse. Organographs are an innovative approach to hierarchical data management.These three contributions are centered on the basic idea of building and sharing hierarchical organizations. Part of these contributions was validated by case studies and, in the case of organographs, an actual implementation.

show abstract

Section: Related Workmentioning

confidence: 99%