Monday, October 15, 2007
Dinosaurs-Let's know DB2 V8
This article is from the Database Expert Sindhujaa Vaageeswaran .
Hi All
Welcome to the world of DB2 V8.
Now that we are in the process of migrating to DB2 V8, let’s start this session with the new features in V8.Though there are 10 significant features identified by the analysts, we shall concentrate on the tips which will be more relevant to the application developers.
As a first step many SQL limitations have been broken in V8.
Naming conventions for tables, views, synonym, function etc has been increased from 18 to 128 bytes.
Name of a column has been increased from 18 to 30 bytes.
Name of cursor created with DECALRE CURSOR WITH RETURN has been increased from 18 to 30 bytes.
The length of character constant, hexadecimal constant and the graphic string constant has been increased to 32704, 32704 and 32698 respectively.
Multi row fetch. That is more than 1 row (or a row set) can be got through a single fetch which is not possible in earlier versions.
Multi row insert.
The length of a column can be altered to a greater size (but not to a smaller length) without drop/recreate of the table.
§ The data type of a column can be switched within character data types (CHAR, VARCHAR); within numeric data types (SMALLINT, INTEGER, FLOAT, REAL, FLOAT, DOUBLE, DECIMAL); and within graphic data types (GRAPHIC, VARGRAPHIC). But you cannot change character to numeric or graphic; numeric to character or graphic; or graphic to numeric or character.
§ The previous data type changes are permitted even for columns that are part of an index or referenced within a view.
§ The identity column characteristics can be altered without table drop/recreate.
§ A column can be added to the end of an index without making it unavailable. Previously we need to drop/recreate the index.
§ The clustering index for a table can be changed without drop and recreate of the index.
§ Changes can be done to partitioned table spaces and indexes that were previously not allowed. For example, the partitioning index can be dropped, A table can be created without a partitioning index, and add a partition can be added to the end of a table to extend the limit key value, partitions can be rotated , and we can even automatically re-balance the partitions.
We can discuss about the multi fetch statement in the following weeks.
Hi All
Welcome to the world of DB2 V8.
Now that we are in the process of migrating to DB2 V8, let’s start this session with the new features in V8.Though there are 10 significant features identified by the analysts, we shall concentrate on the tips which will be more relevant to the application developers.
As a first step many SQL limitations have been broken in V8.
Naming conventions for tables, views, synonym, function etc has been increased from 18 to 128 bytes.
Name of a column has been increased from 18 to 30 bytes.
Name of cursor created with DECALRE CURSOR WITH RETURN has been increased from 18 to 30 bytes.
The length of character constant, hexadecimal constant and the graphic string constant has been increased to 32704, 32704 and 32698 respectively.
Multi row fetch. That is more than 1 row (or a row set) can be got through a single fetch which is not possible in earlier versions.
Multi row insert.
The length of a column can be altered to a greater size (but not to a smaller length) without drop/recreate of the table.
§ The data type of a column can be switched within character data types (CHAR, VARCHAR); within numeric data types (SMALLINT, INTEGER, FLOAT, REAL, FLOAT, DOUBLE, DECIMAL); and within graphic data types (GRAPHIC, VARGRAPHIC). But you cannot change character to numeric or graphic; numeric to character or graphic; or graphic to numeric or character.
§ The previous data type changes are permitted even for columns that are part of an index or referenced within a view.
§ The identity column characteristics can be altered without table drop/recreate.
§ A column can be added to the end of an index without making it unavailable. Previously we need to drop/recreate the index.
§ The clustering index for a table can be changed without drop and recreate of the index.
§ Changes can be done to partitioned table spaces and indexes that were previously not allowed. For example, the partitioning index can be dropped, A table can be created without a partitioning index, and add a partition can be added to the end of a table to extend the limit key value, partitions can be rotated , and we can even automatically re-balance the partitions.
We can discuss about the multi fetch statement in the following weeks.
# posted by Sundar : 12:47 PM
Subscribe to Posts [Atom]