Use the COUPLE function to define one descriptor for each of two files to be coupled.
This document covers the following topics:
- DESCRIPTOR: Descriptors Used as Basis for Coupling
The DESCRIPTOR parameter defines one descriptor in each file to provide the basis for coupling the files. Subdescriptors or superdescriptors may also be used, or may be defined as or derived from a multiple-value field. The descriptors specified may not be contained within a periodic group, nor be derived from a periodic group. The descriptors can have different names, but must have the same length and format definitions.
- FILES: Files to Be Coupled
FILES specifies the two files to be coupled. The number of each file must be 255 or lower. The files specified may not be currently coupled to each other.
- LPB: Prefetch Buffer Size
LPB specifies the size, in bytes, of the internal prefetch buffer. The maximum value is 32760 bytes. The default depends on the ADARUN LU parameter; ADAINV may also reduce a specified LPB value if the LU value is too small.
- LWP: Work Pool Size
LWP specifies the size of the work pool to be used for descriptor value sorting. The value can be specified in bytes or kilobytes followed by a "K". If no value is specified, the default is 1048576 bytes (or 1024K); however, to shorten ADAINV run time for files with very long descriptors or an unusually large number of descriptors, set LWP to a higher value. To avoid problems with the sort data set, a smaller LWP value should be specified when defining descriptors for relatively small files.
The minimum work pool size depends on the sort data set's device type:
Sort Device Minimum LWP Minimum LWP Bytes Kilobytes 3380 139264 136K 3390 159744 156K - NOUSERABEND: Termination without Abend
When a parameter error or a functional error occurs while this utility function is running, the utility ordinarily prints an error message and terminates with user abend 34 (with a dump) or user abend 35 (without a dump). If NOUSERABEND is specified, the utility will not abend after printing the error message. Instead, the message "utility TERMINATED DUE TO ERROR CONDITION" is displayed and the utility terminates with condition code 20.
Note:
When NOUSERABEND is specified, we recommend that it be specified as the first parameter of the utility function (before all other parameters). This is necessary to ensure that its parameter error processing occurs properly.- PASSWORD: File Password
If one or both of the files being coupled is security protected, a valid password for the file (or files) must be specified with this parameter. If both files are password-protected, both must have the same password.
- SORTDEV: Sort Device Type
ADAINV uses the sort data set to sort descriptor values. The SORTDEV parameter indicates the device type to be used for the sort data set. See the z/OS job control information for specific SORTDEV considerations. Refer also to Enhancements to TEMP and SORT Processing in ADAINV, ADALOD and ADAULD for more details on specifying the SORDEV parameter.
- SORTSIZE: Sort Size
SORTSIZE specifies the space available for the sort data set or data sets R1/2. The value can be either cylinders (a numeric value only) or blocks (a numeric value followed by a "B"). If blocks are specified, they should be equivalent to a full number of cylinders.Refer to the Adabas DBA Reference documentation for more information on estimating the sort space. Refer also to Enhancements to TEMP and SORT Processing in ADAINV, ADALOD and ADAULD for more details on specifying the SORTSIZE parameter.
- TEMPDEV: Temporary Storage Device Type
ADAINV uses the temp data set to store intermediate data. The TEMPDEV parameter indicates the device type to be used for this data set. Refer to Enhancements to TEMP and SORT Processing in ADAINV, ADALOD and ADAULD for more details on specifying the TEMPDEV parameter.
- TEMPSIZE: Temporary Storage Size
TEMPSIZE defines the space available for the temp data set. The value may be in cylinders (a numeric value only) or blocks (a numeric value followed by a "B"). Refer to Enhancements to TEMP and SORT Processing in ADAINV, ADALOD and ADAULD for more details on specifying the TEMPSIZE parameter.
- TEST: Test Syntax
The TEST parameter tests the operation syntax without actually performing the operation. Only the syntax of the specified parameters can be tested; not the validity of values and variables.
ADAINV COUPLE FILES=3,4,DESCRIPTOR='AA,BB'
Files 3 and 4 are to be coupled. Descriptor AA from file 3 and descriptor BB from file 4 are to be used as the basis for the coupling.
An intermediate data set is generated for each of the files being coupled.
An entry is written to the data set for each record contained in the file. Each entry contains the ISN of the record (3 or 4 bytes, depending on the ISNSIZE defined for the files) and the value (in compressed form) of the descriptor being used as the basis for the coupling. If the descriptor is defined with the NU option, no entries are written for records in which the descriptor contains a null value. If the descriptor is a multiple-value field, an entry is written for each different value.
The space required for each of the intermediate data sets is a function of the number of records contained in each Adabas file, and the length and the number of the different values present for the coupling descriptor in each record.
Use the following equation to determine the space needed for an intermediate data set:
SP = RECS x UV x (ISNSIZE + (AVLEN x 4)
where
| SP | intermediate data set space required (in bytes). |
| RECS | number of records contained in the coupled file. |
| UV | average number of unique values per record for the descriptor. If the descriptor is not defined with the NU option, UV is equal to or less than 1. If the descriptor is defined with the NU option, UV is equal to the average number of values per record minus the percentage of records that contain a null value. For example, if the average number of values per record is 1 and 20 percent of the values are null, UV is equal to 1 - 0.2 = 0.8. |
| ISNSIZE | length of ISNs in the file (3 or 4 bytes). |
| AVLEN | average length (after compression) of each value for the descriptor. |
The file being coupled has 3-byte ISNs and contains 50,000 records. The descriptor being used as the basis for coupling contains 1 value per record (with no null values) and has an average value length of 5 bytes.
SP = 50,000 x 1 x (3 + (5 + 1)) SP = 50,000 x 9 SP = 450,000 bytes
ADAINV matches the two lists, sorts each resulting list, and writes each list to the Associator coupling lists.
The temp data set stores the matched (coupled) ISNs for each file. An entry is written to the temp data set for each match found. The entry contains the ISN of each record containing a matching value.
ADAINV sorts the entries stored on the temp data set using the sort area and writes the sorted entries to the Associator coupling lists for file A. The same process is then repeated for file B.
The temp area size requirement depends on the number of matching values in the two files for the descriptor used to couple the files. Each match requires 6 or 8 bytes, depending on the ISNSIZE defined for the files.
The sort area generally requires twice the amount of space as that needed for the temp area.
File coupling is bidirectional rather than hierarchical in that two coupling lists are created with each list containing the ISNs which are coupled to the other file.
Assume that 2 files containing the descriptors AA and BB, respectively, are to be coupled. The values for the first five records of each file are as follows:
| File A | File B | ||
|---|---|---|---|
| ISN | Field AA value | ISN | Field BB value |
| 1 2 3 4 5 |
20 25 27 30 40 |
1 2 3 4 5 |
18 40 25 20 20 |
If the two files were coupled using AA and BB as the basis for the coupling, the resulting coupling lists would be:
| File A | File B | ||||
|---|---|---|---|---|---|
| ISN in FILE B* | COUNT | COUPLED ISNs | ISN in FILE A* | COUNT | COUPLED ISNs |
| 2 3 4 5 |
1 1 1 1 |
5 2 1 1 |
1 2 5 |
2 1 1 |
4,5 3 2 |
* Internally, Adabas uses this field like a descriptor to determine the number and the ISNs of the coupled records.
The total space requirement for the coupling lists depends upon the number of common values that exist between the two descriptors used as the basis for the coupling.
The space requirement for each common value may be estimated as follows:
SP = 4a + 4b + 6ab
where
| SP | space requirement for one common value (in bytes); |
| a | number of records in file A containing the common value; |
| b | number of records in file B containing the common value. |
The total coupling list requirement is the sum of the space requirements of each common value.
Using sample files A and B as previously defined, space requirements per common value are
| Common Value | Space Requirements |
|---|---|
| 20 | SP = 4(1) + 4(2) + 6(1 • 2) = 24 bytes |
| 25 | SP = 4(1) + 4(1) + 6(1 • 1) = 14 bytes |
| 40 | SP = 4(1) + 4(1) + 6(1 • 1) = 14 bytes |
Total space required = 24 + 14 + 14 = 52 bytes
Assume that 2 files are being coupled on the field ID. The values for ID are unique within each file. There are 5,000 common values in the coupled files.
| Common Value | Space Requirements |
|---|---|
| n | SP = 4(1) + 4(1) + 6(1) SP = 14 bytes for one common value |
There are 5,000 common values, each of which requires 14 bytes. The total space requirement for the coupling lists is 70,000 bytes.
The coupling lists constructed by ADAINV are contained within the normal (NI) and upper (UI) index for each file being coupled. If the NI or UI component's logical extents currently allocated to the file are used up during ADAINV execution, ADAINV attempts to allocate an additional extent to the component. The size of the extent allocated is equal to 25 percent of the current total size of all logical extents currently assigned to the component. If insufficient space is available or if the maximum number of allocated extents has been reached for the component, ADAINV terminates with an error message.
The TEMPSIZE and SORTSIZE parameters are now optional. When omitted, various enhancements to TEMP/SORT processing are available to the user. See Enhancements to TEMP and SORT Processing in ADAINV, ADALOD and ADAULD for details.