Review of Materialized Views Selection Algorithm for Cyber Manufacturing

Abstract

Technological advancement in data transfer and connection has driven massive data growth. Within the semiconductor cyber manufacturing environment, in order to cope with rapid data transfer enabled by the Internet of Things (IoT) technology, rapid query processing becomes a priority. Especially, in the era of Industry 4.0, semiconductor manufacturing that operates within cyber-physical systems (CPS) relies heavily on the reporting function to monitor delicate wafer processing. Thus, delay in reporting which is usually caused by slow query processing is intolerable. Materialized views (MVs) are usually used in order to improve query processing speed. Nevertheless, as MVs requires database space and maintenance, the decision to use MVs is not determined by time factor only. Thus, MVs selection is a problem that calls for an efficient selection algorithm that can deal with several constraints at a time. In this paper, we reveal the criteria of optimisation algorithms that were proposed to deal with MVs selection problem. In particular, this paper attempts to evaluate the coverage and limitations of the algorithm under study.