@misc{Tuschl, type = {Master Thesis}, author = {Tuschl, Roland}, title = {Ein Userspace-Dateisystem zur Verwaltung von Grid-Daten}, url = {http://nbn-resolving.de/urn:nbn:de:0297-zib-11483}, school = {Zuse Institute Berlin (ZIB)}, abstract = {Das Ziel dieser Arbeit ist die Schaffung einer Zugriffs-Komponente f{\"u}r das Grid-Datenmanagement-System ZIB-DMS, das dessen transparente Einbindung in den Verzeichnisbaum eines Linux-Systems erlaubt. Dazu wird unter Verwendung des FUSE-Rahmenwerkes ein Userspace-Dateisystem mit Anbindung an das ZIB-DMS konzipiert und implementiert. Im Fokus stehen dabei die Abbildung der erweiterten Verwaltungsmechanismen des Systems auf die limitierte Schnittstelle hierarchischer Dateisysteme und die dazu notwendigen {\"A}nderungen am ZIB-DMS.}, language = {de} } @misc{SchintkeReinefeld, author = {Schintke, Florian and Reinefeld, Alexander}, title = {On the Cost of Reliability in Large Data Grids}, doi = {10.1023/B:GRID.0000024086.50333.0d}, url = {http://nbn-resolving.de/urn:nbn:de:0297-zib-7195}, number = {02-52}, abstract = {Global grid environments do not only provide massive aggregated computing power but also an unprecedented amount of distributed storage space. Unfortunately, dynamic changes caused by component failures, local decisions, and irregular data updates make it difficult to efficiently use this capacity. In this paper, we address the problem of improving data availability in the presence of unreliable components. We present an analytical model for determining an optimal combination of distributed replica catalogs, catalog sizes, and replica servers. Empirical simulation results confirm the accuracy of our theoretical analysis. Our model captures the characteristics of highly dynamic environments like peer-to-peer networks, but it can also be applied to more centralized, less dynamic grid environments like the European {\em DataGrid}.}, language = {en} } @misc{Best, author = {Best, Christoph}, title = {A simple C++ library for manipulating scientific data sets as structured data}, url = {http://nbn-resolving.de/urn:nbn:de:0297-zib-5674}, number = {TR-98-06}, abstract = {Representing scientific data sets efficiently on external storage usually involves converting them to a byte string representation using specialized reader/writer routines. The resulting storage files are frequently difficult to interpret without these specialized routines as they do not contain information about the logical structure of the data. Avoiding such problems usually involves heavy-weight data format libraries or data base systems. We present a simple C++ library that allows to create and access data files that store structured data. The structure of the data is described by a data type that can be built from elementary data types (integer and floating-point numbers, byte strings) and composite data types (arrays, structures, unions). An abstract data access class presents the data to the application. Different actual data file structures can be implemented under this layer. This method is particularly suited to applications that require complex data structures, e.g. molecular dynamics simulations. Extensions such as late type binding and object persistence are discussed.}, language = {en} }