SQL Processing - Relational Database Index Design and the Optimizers - page 29

Chapter 3

SQL Processing

ž How indexes and tables are used to process SQL statements
ž SQL processing concepts
ž Predicates
ž Optimizers and access paths
ž Index slices, matching index scans, and matching columns, index
screening, and screening columns
ž When the access path is chosen
ž Monitoring the optimizer
ž Helping the optimizer with statistics and the number of FETCH
calls required
ž Concepts of filter factor, selectivity, and cardinality, together with their
impact on index design
ž The materialization of the result table and its implications

INTRODUCTION

We now have some understanding of the structure of tables and indexes; we also
understand how these objects relate to buffer pools, disks, and disk servers and
how the latter are used to make the data available to the SQL process. We are
now in a position to consider the processing of the SQL calls.
As before, much will depend on the individual relational DBMS being used;
here again, there are many similarities, but there are also a number of differences.
We will describe the underlying processes first in general terms in order to avoid
confusing the basic issues with DBMS specifics; the latter will be addressed
as appropriate. Considerably more detail of these processes will be provided
throughout the topic, at a time when it is more convenient to do so. Remember
that the glossary provided at the end of the topic summarizes all the terms used
throughout this chapter.