TurboIMAGE/XL Database Management System Reference Manual (30391-90012)

Appendix C 651
Database Design Considerations
C Database Design Considerations
• Keep one-of-a-kind information, such as unique identifiers, in master data sets. Keep
duplicate information, such as records of events (sales, purchases, shipments), in detail
data sets.
• Define a search item in a detail data set if you want to retrieve all entries with a
common value for that data item.
• Use manual master data sets to prevent entry of invalid data in the detail search item
linked to the master through a path.
Use automatic master data sets to save time if the detail search items are unpredictable
or too numerous to enter manually.
• Limit the use of sort items to paths with relatively short chains in order to reduce the
time required to add and delete entries.
• Select the path most frequently accessed in chained order as the primary path.
• Remember that data items must be an integral number of halfwords in length.
• When selecting the maximum block size, consider the environment in which the
database will be used. (Refer to the $CONTROL command in chapter 3 for more
information.)
• If you intend to use QUERY/3000 with your database, refer to the QUERY/V Reference
Manual for the data types that QUERY/3000 supports.
• In application programs either reference data items and data sets by name or use
DBINFO at the beginning of the program to initialize the data item and data set
numbers in order to maintain data independence of the programs.
• Refer to the discussion in chapter 4 to decide on appropriate access modes to use for
your application programs.
• Analyze the time required to maintain the database, for example, the time required to
unload and load the database. A database restructuring tool can help keep the time
spent on maintenance to a minimum.
• The capacity of each data set should be defined as realistically as possible because a
capacity that is too large wastes disk space. Data set can be expanded dynamically if
capacity expansion parameters are set. Otherwise, the capacity can be increased when
necessary by restructuring the database as described in chapter 8.
• A master data set capacity equal to a prime number or to the product of two or three
primes can yield fewer synonyms than a master data set capacity of many prime
factors. See Figure C-1. for a partial list of prime numbers.
• The account and group in which the database resides must have enough file space
available to contain all the database files.
• If your application uses sorted paths, plan to add or delete entries (DBPUT, DBDELETE)
to sorted chains when the system is not very busy. If it is very busy, limit the database