Creating a workload definition
You can use workload definitions to route work requests to a specific set of target regions based on the terminal and user names, or the process types, associated with those work requests. This section describes how to create a workload definition and add it to the data repository.
The terminal and user names may be either specific or generic. For example, you can create a workload definition that causes all transactions initiated by any user from terminals with logical unit names starting with NET to be routed to the target scope identified as EYUCSG01.
- Is the transaction identifier part of a transaction group associated with the workload definition?
- Is there a definition specifying separation by process type? Is there a match? If so, use this definition. If not, is there a definition specifying separation by terminal logical unit name and user ID?
- In what order should the terminal and user names associated with the transaction be evaluated; that is, which name is to be used as the primary filter?
- Do the user and terminal names associated with the transaction match the name patterns specified with the workload definition?
- When there are multiple workload definitions, the field containing
the name identified as the primary filter can contain the same specific
or generic pattern; the contents of the field used as the secondary
filter must always be unique. For example, the following definitions
are valid when USERID is the primary filter and LUNAME is the secondary
filter. They are not valid, however, when LUNAME is the primary filter
because the user identifiers are not unique.
Entry Name : EYUWMD01 EYUWMD02 EYUWMD03 Trangroup Name : EYUTNG01 EYUTNG01 EYUTNG01 Terminal Luname: LUR* LUL* LUT* User ID : PIE* PIE* PIE* - When a generic name is specified, the pattern with the most matching
leading characters has precedence. For example, with the following
patterns:
Pattern A37AR* is selected when the input is A37AR123. Pattern A37+R* is selected when the input is A37TRAP.A37AR* A37+R* - Terminal LU name patterns are matched on the concatenated values
of network name and logical unit name. The following are valid terminal
LU name patterns:
NETWORK1.LU12345678 - A specific luname NETWORK1.LU1* - Generic lunames in the network NETWORK1.* - All lunames in the network NET* - All lunames in a generic network .LU12345678 - A specific luname within all networks .* or * - All lunames within all networks
To create a workload definition and add it to the data repository:
- Click to open the Workload management
definition view.
This view displays a list of existing workload management definitions. It has action buttons that allow you to create, update, remove and install workload definitions, and to add a workload definition to a workload group.
- If you want to use some of the information from an existing definition in the creation of your new definition, select an existing definition by selection by selecting an adjacent check box in the Record column.
- Click the Create action button.
- Provide the following information, as appropriate:
- Workload management definition name
- Specify a 1- to 8-character name for the workload definition. The name can contain alphabetic, numeric, or national characters. However, the first character must be alphabetic.
- Description
- (Optional) Specify a 1- to 30-character description of the definition.
- Trangroup group name
- (Optional) Enter the specific or generic name of a transaction
group. If you enter a generic value, a list of valid transaction groups
is displayed.
If you do not identify a transaction group, the default transaction group for the specification is assumed.
- Terminal LU name
- Enter a specific logical unit name or a pattern, using the format:
where both the network_qualifier_name and the luname are between 1 and 8 characters in length. Both the network_qualifier_name and the luname are optional, however one must be specified. If the luname is specified it must be prefixed by a period sign.[network_qualifier_name][.luname]A pattern can include the characters + (plus sign), * (asterisk), or both; where the plus sign represents any non-blank character and can be specified anywhere in the pattern and the asterisk represents a range of zero or more characters that may include blanks. The asterisk must only be used at the end of the pattern.
- User ID
- Enter a specific user ID or a pattern. A pattern can include the
characters + (plus sign), * (asterisk), or both. Note: The Terminal Luname and User ID values are used when CICSPlex SM attempts to match a transaction with this definition.
- BTS process type
- Enter a specific process type or a pattern. A pattern may include
+ (plus sign), * (asterisk), or both. Note: If you specify anything other than an * in this field, the luname and user id fields must be set to an *. Similarly, if there is anything other than an * in either luname and user id, this field must be set to *. You cannot separate a workload by process type, luname and user id.
- Scope name of set of target systems
- Enter
the specific or generic name of a CICS® system
or CICS system group to be used as target regions
for dynamic routing requests. If you enter a generic value, a list
of valid CICS systems and CICS system
groups is displayed. Note: The specified CICS system or CICS system group must be accessible to the CICS systems identified as the routing scope with the workload specification to which this definition is associated. Therefore, you should verify that the appropriate CICS connections exist between each routing region and all of the CICS systems identified here as the target scope.
- Click Yes to create the new definition.
The Workload management definition view is redisplayed and includes an entry for the new definition.