Projects

Abstract: RadegastXDB (ver. 0.3) is a prototype of semi-structured database management system, that implements the XQuery query language. We include state-of-the-art query processing algorithms and optimalization techniques. RadegastXDB is build on top of the QuickDB data store, therefore, it is persistent and uses only preallocated memory.

Description: There are successfull builds for the Microsoft Windows (32bit and 64bit) and Linux (32bit and 64bit) platforms. RadegastXDB uses the Xerces XML parser (http://xerces.apache.org/) (ver. 3.1). Libraries of the parser are included in the ZIP archive below.The archive also contains examples of configuration XML files to index and query 4 data collections: XMark, Swissprot, DBLP and TreeBank. Testing query sets are also included.

Abstract: QuickDB is a embeded persistent data storage implementing several basic data structures such as persistent array (heap), B-tree, Hash map and R-tree. It is oriented on a performance of basic operations such as insert, point query and range query. It contains its own caching mechanism and memory pool, therefore, no memory allocations occur without explicit request. The API of QuickDB is similar to other embeded data storages such BerkelyDB. Therefore, there is no SQL layer and data storage has to be instantiated and used through a set of classes and dedicated methods.

Description: There are currently succesfull builds on a windows and linux OS.

Abstract: This library includes encoding/decoding functions of conventional as well as fast algorithms for the following codes: Elias-delta, Fibonacci 2 and 3, and Elias-Fibonacci. This library follows the article:J. Walder, M. Krátký, R. Bača, J. Platoš, and V. Snášel. 2012. Fast decoding algorithms for variable-lengths codes. Inf. Sci. 183, 1 (January 2012), 66-91. http://dx.doi.org/10.1016/j.ins.2011.06.019

Description: The archive file includes a sample and the library for Visual Studio (both Win32 and x64 platforms).