Abstract:
This paper presents an implementation of a novel
approach, utilizing hybrid data partitioning, to secure sensitive
data and improve query performance. In this novel approach,
vertical and horizontal data partitioning are combined together
in an approach that called hybrid partitioning and the new
approach is implemented using Microsoft SQL server to
generate divided/partitioned relations. A group of proposed
rules is applied to the query request process using query binning
(QB) and Metadata of partitioning. The proposed approach is
validated using experiments involving a collection of data
evaluated by outcomes of advanced stored procedures. The
suggested approach results are satisfactory in achieving the
properties of defining the data security: non-linkability and
indistinguishability. The results of the proposed approach were
satisfactory. The proposed novel approach outperforms a wellknown approach called PANDA.
