Tasks that LOADPLUS performs

Resources

• Analyzes and allocates memory and CPU resources for maximum throughput
• Optionally analyzes data set resources that are needed for the specific load job
• Optionally offloads eligible processing to an IBM zIIP
• Optionally estimates input file size automatically for dynamic allocation and efficient sort processing
• Optionally allocates input, load data, index work, sort work, discard, error, and copy files dynamically

Indexes

• Optionally reorganizes the participating indexes
• Optionally updates indexes, which can improve performance when loading a small number of rows into an existing table that has a large number of rows
• Optionally skips building or updating nonpartitioned indexes

Performance

• Optionally provides single-phase load processing, which combines the PRELOAD and LOAD phases into one phase (the COMBINED phase)
• Provides an option to order data by table and clustering key or by clustering key alone
• Optionally defers checks for duplicate keys in unique indexes
• Provides an option to improve performance by not sorting the data or clustering index if the data is already in the correct clustering index sequence
• Provides an option to improve performance when transferring data between similarly defined tables by loading data that was unloaded by the UNLOAD PLUS in an internal format

Object availability

• Provides an option for the utility to pause in its processing before the data is loaded if input records are discarded
• Runs concurrently with other BMC utilities on Db2 spaces
• Provides an option to leave tables in RO status while loading your data
• Provides an option to leave tables in RW status while loading your data

Compression

Builds or keeps a compression dictionary when the table space or partition being loaded is defined with the COMPRESS YES attribute

Loading data

• Loads data from one or more data sets into one or more tables or table space partitions in a single command execution, either replacing or adding to the existing data
• Optionally sorts the input data by the clustering key or partitioning key
• Optionally orders rows by table
• Allows you to load data that was archived by the REORG PLUS product
• Allows you to load data that was unloaded by the UNLOAD PLUS product in an internal format
• Allows you to load data from batch pipes
• Provides powerful selection criteria for specifying which input records to load
• As part of the load, optionally deletes and redefines user-defined data sets and data sets that are defined in DB2 storage groups

• Avoids adding rows and keys to the table spaces and indexes if they cause duplicate keys in a unique index, so that the rows and keys do not need to be deleted later in the LOAD phase

  Following this process, your table spaces and indexes remain organized.

Data integrity

Optionally checks the following constraints:

• Db2 defined check constraints
• Db2 defined referential constraints
• Informational referential constraints

Copies

• With a self-contained copy function, produces image copies or DSN1COPY-type copies concurrently with the load
• Optionally dynamically allocates the image copy data sets
• Optionally produces inline image copies while loading your tables, rather than DSN1COPY-type copies or image copies after loading your tables

Statistics and reports

• Optionally updates statistics in the Db2 catalog so that the Db2 optimizer can use them
• Optionally updates statistics in the DASD MANAGER PLUS product's database statistics tables
• Produces a statistics report in the SYSPRINT output or, optionally, in a separate data set
• Updates the Db2 real-time statistics tables
• Prevents the output of unnecessary discard messages and records by allowing you to ignore specified types of discards
• Reports multiple load command errors at one time, reducing the number of runs needed to correct your load specification

Operations on data

• Provides a comprehensive set of data type conversions and support for user-written exit routines that perform special data conversions
• Provides options to perform data translation
• Optionally provides a PRELOAD phase for verifying that your input data is correct before your existing data is replaced (and lost), without incurring the overhead of actually loading the data
• Provides full Boolean logic on the WHEN condition (allows AND, OR, NOT, and parentheses)
• Provides control over expressions used in WHEN, NULLIF, and DEFAULTIF processing through the RULES installation option
• Allows you to concatenate separate physical input records into one larger logical record
• Allows you to assign a constant value or CURRENT DATE, CURRENT TIME, or CURRENT TIMESTAMP to a column
• Allows multiple NULLIF and DEFAULTIF conditions on a field specification
• Provides powerful criteria for assigning NULL or default values to columns:
  • Allows you to assign default values to nullable or nondefaultable columns
  • Allows you to specify the value for a DEFAULTIF value, overriding the standard default
  • Allows you to assign NULL or a default value to a column if a conversion error occurs while LOADPLUS is processing the column’s input value

Job and object status

• Provides restart capabilities
• Optionally allows you to set user-defined return codes for a range of situations
• Sets the status of all affected table spaces to CHECK pending (CHKP) if referential integrity constraints or check constraints exist and were not checked