Database Reference
In-Depth Information
Chapter 4
Compression Schemes of High
Dimensional Data for MOLAP
K. M. Azharul Hasan
Khulna University of Engineering and Technology (KUET), Bangladesh
ABStrAct
The exploration of the possibility of compressing data warehouses is inevitable because of their non-trivial
storage and access costs. A typical large data warehouse needs hundreds of gigabytes to a terabyte of stor-
age. Performance of computing aggregate queries is a bottleneck for many Online Analytical Processing
(OLAP) applications. Hence, data warehousing implementations strongly depend on data compression
techniques to make possible the management and storage of such large databases. The efficiency of data
compression methods has a significant impact on the overall performance of these implementations. The
purpose of this chapter is to discuss the importance of data compression to Multidimensional Online
Analytical Processing (MOLAP), to survey data compression techniques relevant to MOLAP, and to
discuss important quality issues of MOLAP compression and of existing techniques. Finally, we also
discuss future research trends on this subject.
IntroductIon
Reducing the storage requirement of databases is
equivalent to increasing the capacity of the storage
medium. Since compressed data are encoded using
a smaller number of bytes, transfer of compressed
information from one place to another requires less
time and hence results in a higher effective transfer
rate. Since data compression reduces the loading of
I/O channels, it becomes feasible to process more
I/O requests per second and hence achieve higher
effective channel utilization. Most importantly,
however, is the application of data compression
Data compression is widely used in data manage-
ment to save storage space and network bandwidth.
The main benefits that are achieved in data com-
pression are well described by Bassiouni 1985
for different contexts and applications. The most
obvious advantage of data compression is that of
reducing the storage requirement for the database.
Search WWH ::
Custom Search