Knowledge and Cache Conscious Algorithm Design and Systems Support for Data Mining Algorithms

The knowledge discovery process is interactive in nature and therefore minimizing query response time is impera- tive. The compute and memory intensive nature of data mining algorithms makes this task challenging. We pro- pose to improve the performance of data mining algorithms by re-architecting algorithms and designing effective sys- tems support. From the view point of re-architecting algo- rithms, knowledge-conscious and cache-conscious design strategies are presented. Knowledge-conscious algorithm designs try and re-use repeated computation between iter- ations and across executions of a data mining algorithm. Cache-conscious algorithm designs on the other hand re- duce execution time by maximizing data locality and re- use. The design of systems support that allows a variety of data mining algorithms to leverage knowledge-caching and cache-conscious placement with minimal implementa- tion efforts is also presented.