Publication at Faculty of Mathematics and Physics |
2010
Abstract
Effective processing of location axes is essential for fast XPath implementation. Direct approach can be quite inefficient and the performance of the whole XPath processor may suffer, since location axes are present in every location step of every location path.
We propose axes optimizations, indices and algorithms that should improve performance in many situations. These algorithms are designed to accomodate features of nowadays processors and they can be implemented in parallel fashion without extensive modifications.