Database Utilities › ADDDS

ADDDS

The ADDDS control statement is used to create a new database, to add additional space to an existing database, or to add a new index file data set. Space is added by creating a new data set and formatting it with fixed-length blocks.

Database Extent Considerations

• For data sets that are 4,369 or fewer cylinders, the data sets are to be allocated as an IBM direct access data set.
• For data sets that are more than 4,369 cylinders, the data sets are to be allocated as physical sequential data sets due to IBM restrictions.
• z/OS V1R5.0 introduced Enhanced Data Integrity (EDI). EDI allows customers to protect against concurrent access of shared sequential data sets for update or output. The CA View database should be exempt from the EDI function by adding the CA View data set names to the EDI exclusion list (found in SYS1.PARMLIB(IFGPSEDI)). CA View has its own internal protection mechanism so that EDI protection is not needed.

Note: The physical sequential data sets are accessible via standard utilities and ISPF. It is recommended that you use the CA View Systems Extension Data Set Security Feature to limit access to these data sets or implement encryption to secure the data.

To minimize contention, we recommend that each database extent be placed on a separate dedicated volume. Where possible, the size of the volume should match the size of the database extent. A matching database extent prevents I/O for multiple extents from queuing on the same device address.

Also, the first extent of the index and the database must be placed on separate dedicated volumes because the RESERVE processing uses the dataset name and volume of the first extent to serialize all accesses.

The high-level name of the database must have previously been defined with the NAME control statement (or the PARM parameter of the EXEC JCL statement).

Note: The allocation of a new index and/or data extent might not take effect immediately if archival tasks are active. (These archive tasks are SARSTC, FSS Collectors, or CA Deliver Direct-to-View.) Be aware of the following:

• The archival task does not need to be recycled for either a new index or data extent allocation to take effect.
• The new extent becomes active to an archival task when that processor makes a space allocation request.
• A newly allocated data extent becomes active to an archival task when a new SYSOUT is archived
• A newly allocated index extent becomes active to an archival task when the index is expanded by an index space allocation (cylinder allocation) request.

  Syntax

  	ADDDS	ABOVE
  		BLKSIZE=
  		CYLINDER=
  		DATACLASS=
  		INDEX|DATA
  		MGMTCLAS=
  		STORCLAS=
  		UNIT=
  		VOLSER=
  

Where:

ABOVE

Indicates that the data set being added to the database can be allocated above the 64K cylinder line on a 3390-A device. Data sets that are allocated above the 64K cylinder line are allocated in increments of 21 cylinders so the final data set allocation is rounded up to the next multiple of 21 cylinders. This parameter is optional.

BLKSIZE=

Specifies the block size to be used in the data sets.

The minimum is 3476; the maximum is 32760. The default block size for the database index file data sets is 8906, and the default block size for the database data file data sets is 3768. The block size for an existing database cannot be changed, so this operand is only valid when you create a new database. After a database has been created, any subsequent use of the block size parameter is ignored.

Note: For more information about determining the block size to use, see Estimating DASD Requirements For a Database later in this chapter.

CYLINDER=

Specifies the number of contiguous cylinders to allocate to the data set. A maximum of 32,760 cylinders can be allocated to one data set.

DATACLASS=

Specifies the data class of an SMS dataset

INDEX|DATA

Specifies whether an index file or a data file data set is to be added to the database. INDEX indicates the addition of an index file data set suffixed with Innnnnnnn. DATA indicates a data file data set suffixed with Dnnnnnnnn. Either INDEX or DATA is required.

MGMTCLAS=

Specifies the management class of an SMS data set

STORCLAS=

Specifies the storage class of an SMS data set

UNIT=

Specifies the unit name to be used to dynamically allocate a new data set

VOLSER=

Specifies the volume serial number on which to allocate the new data set

All of the ADDDS keywords can be abbreviated to the fewest number of characters that makes them unique.

SMS and Non-SMS: Specifying Parameters

The following table summarizes the interdependency of the ADDDS parameters for SMS and non-SMS data sets:

Estimating DASD Requirements for a Database

There is no way to know the exact amount of space that a database might require.

These formulas can be used to approximate the space needed (for this approximation, a block size of 3768 is used):

• If you are installing a new database, you must allocate 100 cylinders of INDEX for every 45,000 reports, including reports on tape. This allocation gives a 50 percent buffer to do standalone reorganization.
• CA View writes fixed-length blocks to its database. (You can specify a block size from 3476 through 32760.)
• For a block size of 3768, 195 blocks are contained in one cylinder of a 3390, 165 blocks are contained in one cylinder of a 3380, and 130 blocks are contained in one 3350 cylinder.
• Repetitive characters are compressed, giving, on the average, a 60 to 70 percent reduction in space requirements.
• The space required for the master index is minimal compared to the space required for the SYSOUT data; a minimum of one cylinder is required for the master index.

Procedure to Estimate Database Space

Using the previous information, you can use this procedure to approximate the amount of space required for your database.

1. Take the maximum lines archived to disk in one generation (that is, one day).
2. Multiply the number of lines by the average line length.
3. Divide by 3 (for compression).
4. Divide by 3768 to get number of blocks.
5. Divide by 195 (for a 3390) or 165 (for a 3380) or 130 (for a 3350) to get number of cylinders.
6. Multiply the result by the number of generations to be retained on disk.
7. To allow for growth, add 10 percent to the result.

Example

Assume that a site produces 1,000,000 lines of SYSOUT in one day to be archived on disk. The average line length of the SYSOUT is 121 bytes, and three days' worth of SYSOUT is going to be kept on disk.

Using the method described on the previous page, space requirements can be approximated as follows:

• 1,000,000 lines (per generation)
• 1,000,000 lines * 121 bytes/line = 121,000,000 bytes
• 121,000,000 bytes / 3 = 40,333,333 bytes (compressed)
• 40,333,333 bytes / 3768 bytes/block = 10,705 blocks
• A block size of 3768 is assumed for this example. Consult your DASD administrator to choose an optimal block size.
• 10,705 blocks / 195 3390-cylinders = 55 3390-cylinders
• 55 3390-cylinders * 3 generations = 165 3390-cylinders
• 165 3390-cylinders + 17 3390-cylinders (10%) = 182 3390-cylinders