Databases Reference
In-Depth Information
table. All the rows constitute a set, specifically a set of rows. The RDBMS
retrieves all the rows from the table and delivers them to the user. If a
query selects data from two tables, then it retrieves two sets of rows as it
builds the final result set. If a query retrieves data from three tables, then
it retrieves three sets of rows as it builds the final result set. If a table has
an index on a column, and a query uses a WHERE clause on the specific
column, the RDBMS will use the index to retrieve only the subset of rows
that match the WHERE clause in the query as it builds the final result set.
That subset of rows is still a set of rows. So, no matter how you write the
query or index the tables, the RDBMS will always retrieve data in sets of
rows. This is the set logic inherent in every RDBMS.
Set logic is a very powerful tool and an innovative leap forward past the
flat file and hierarchical databases that preceded relational data structures.
Prior to relational data structures, data was stored in flat files or keyed
files. A procedural application was required to retrieve data from these
files and databases. A business user could not submit an ad hoc query
against a flat file, join that with a keyed file, and then join the result set
of those two files with another keyed file. Prior to relational data struc-
tures, that technology did not exist. That technology is the relational data
structures of an RDBMS. So, the set logic of an RDBMS is not a weakness.
Rather, it is an innovative leap forward that suddenly made large volumes
of data available to business users. Instead of waiting for an application to
run a job that would create a file that was a combination of the data in two
separate files, the creation of an RDBMS gave business users the ability to
retrieve the data they need without waiting for a job to run a program that
reads the files, combines their data, and then writes a file.
Set logic give an RDBMS the ability to manipulate large volumes of data,
join separate sets of data, and combine their data into a single result set
without a job running a program that generates a file. Instead, the busi-
ness user submits a query and the RDBMS does all the work. When the
RDBMS is finished, it passes the result set to the business user who sub-
mitted the query. For that reason, set logic is the strength of every RDBMS.
sets—the BANe of tIme vArIANce
Set logic, the strength of every RDBMS, is the bane of Time Variant data. The
problem is the set of rows. All the millions of occurrences of a transaction
