Tables

Table access

For data-intensive applications, application speed is mainly based upon the number of records being read from and written to tables. The more records that are read and written, the slower the application. To improve the speed of an application, minimize the number of records that are read from and written to tables.

Clearing tables

If you are removing all records from a table, consider deleting and then recreating the table. For tables containing many records, this may be faster than removing the records individually or setting up a range for the entire table and removing the range.

Keys

The time required to write records to a table is directly proportional to the number of keys defined for the table. To improve performance, reduce the number of keys in each table.

The total length of the segments of a key also affects performance. The longer the segments for the key, the larger the table index becomes. Larger table indexes take longer to search, increasing the time to read and write records. To improve table performance, reduce the number of key segments and the length of the segments for each key.

Ranges

If you are working with groups of records in a table, use ranges when possible. Some database managers, such as SQL database managers, perform operations on ranges of records very efficiently. For example, SQL database managers can remove a range of records much faster than they can remove the records individually.

Table size

As tables become larger, read and write times become longer. To improve table performance, keep the size of tables as small as possible.

Temporary tables

In some cases, you may be able to speed up network operations by using temporary tables. For example, if you’re doing considerable processing on a range of records in a table stored on a network drive, you may want to copy those records to a temporary table stored on your local drive, perform the necessary processing using the temporary table, then copy the records back to the table on the network. This can greatly reduce network traffic and improve performance.

When creating temporary tables, create only one key for the table and keep the length of the key to a minimum. This will improve the performance when reading and writing records for the temporary table.


Documentation Feedback