Table of Contents (exploded view)
Abstract for z/OS MVS Planning: Workload Management
Summary of changes
Changes made in z/OS Version 2 Release 2 (V2R2) as updated March 2017
Changes made in z/OS Version 2 Release 2 (V2R2) as updated December 2015
Summary of changes for z/OS Version 2 Release 2 (V2R2)
Changes made in z/OS Version 2 Release 1 (V2R1) as updated December 2013
Changes made in z/OS Version 2 Release 1
Why MVS workload management?
Problems addressed by workload management
MVS workload management solution for today and tomorrow
What is MVS workload management?
Performance administration
Performance management
Workload balancing
Workload management concepts
What is a service definition?
Why use service policies?
Organizing work into workloads and service classes
Performance goals
Resource requirements
Business importance
Why use resource groups?
Assigning work to a service class
Why use application environments?
Why use scheduling environments?
Summary of service definition and service policy concepts
Workload management participants
Workload management work environments
Subsystem support for goal types and multiple periods
Subsystem-specific performance hints
Workload balancing
Workload management in a CPSM environment
Workload management in a DB2 Distributed Data Facility environment
Batch workload management
Multisystem enclave support
Intelligent Resource Director
HiperDispatch mode
Overview
The concept of HiperDispatch mode
Processor categories
Setting HiperDispatch mode in SYS1.PARMLIB
I/O storage management
Handling service class periods with a response time goal
Handling service class periods with a velocity goal
Handling other I/O requests
Controlling the information passed to the I/O manager
Non-z/OS partition CPU management
Workload management and Workload License Charges
Defining the capacity of a partition
Defining group capacity
Workload management with other products
Setting up a service definition
Specifying a service definition
Storing service definitions
Defining the parts of a service definition
Defining service policies
Using policy overrides
Defining workloads
Defining a departmental workload
Defining resource groups
Calculating an LPAR share — Example 1
Specifying the capacity as a number of CPs — Example 2
Defining service classes and performance goals
Velocity formula
Defining performance goals
Determining system response time goals
Examples of service classes with response time goals
Defining velocity goals
Adjusting velocity goals based on samples included in velocity calculation
Using velocity goals for started tasks
Using discretionary goals
Using performance periods
Defining goals appropriate for performance periods
Using importance levels in performance periods
Defining classification rules
Defining classification rules for each subsystem
Defining work qualifiers
Accounting information
Client accounting information
Client IP address
Client transaction name
Client user ID
Client workstation name
Collection name
Connection type
Correlation information
LU name and netid
Package name
Perform
Plan name
Priority
Procedure name
Process name
Scheduling environment name
Subsystem collection name
Subsystem instance
Subsystem parameter
Sysplex name
System name
Transaction class / job class
Transaction name / job name
User ID
zEnterprise service class name (ESC) from a Unified Resource Manager performance policy
Defining z/OS WLM classification rules for performance policies from the Unified Resource Manager
Defining special reporting options for workload reporting
Defining the order of classification rules
Defining a subsystem service class default
Using inheritance in classification rules
System-provided service classes
Organizing work for classification
Using masking notation
Using wildcard notation
Using the start position
Using groups
Using the system-supplied service classes
Defining report classes
Defining service coefficients and options
Calculating the amount of service consumed
Service definition coefficients
Changing your coefficient values
Using the storage (MSO) coefficient for calculations
Specifying I/O priority management
Considerations for I/O priority management
Enabling I/O priority groups
Considerations for I/O priority groups
Specifying dynamic alias management
Workload management considerations for dynamic alias management
HCD considerations for dynamic alias management
Defining special protection options for critical work
Long-term storage protection
Storage critical for address spaces
Storage critical for CICS and IMS transactions
Long-term CPU protection
Long-term I/O protection
Honor priority
Modifications of transaction response time management
Sample scenarios
Scenario 1
Scenario 2
Scenario 3
Scenario 4
Scenario 5
Scenario 6
Reporting
Option summary
Defining application environments
Getting started with application environments
Specifying application environments to workload management
Selecting server limits for application environments
How WLM manages servers for an application environment
Using application environments
Managing application environments
Using operator commands for application environments
Making changes to the application environment servers
Changing the definition of an application environment
Handling error conditions in application environments
Authorizing application environment servers
Example for restricting access to application environment servers
Defining scheduling environments
Getting started with scheduling environments
Specifying scheduling environments to workload management
Managing resource states
Associating scheduling environments with incoming work
Displaying information about scheduling environments and resource states
MVS operator commands
JES2/JES3 operator commands
SDSF commands
Managing performance with the guest platform management provider (GPMP)
Overview of the guest platform management provider (GPMP)
Configuring the guest platform management provider on z/OS
Before You Begin
About this task
Procedure
Results
What to do next
Workload management migration
Creating a service definition for the first time
Migrating to a new z/OS release with an existing service definition
Migration activities
Restricting access to the WLM service definition
Restricting access to an MVS data set
Restricting access to the WLM couple data set
Start the application and enter/edit the service definition
Before starting the WLM application
Start the WLM application
Enter/edit the service definition
Service definition functionality levels, CDS format levels, and WLM application levels
Calculate the size of the WLM couple data set
Allocate a WLM couple data set
Sample JCL to allocate a WLM couple data set
Increasing the size of the WLM couple data set
Make a WLM couple data set available to the sysplex for the first time
Using the SETXCF command
Updating the COUPLExx member
Make a newly formatted couple data set available to the sysplex
Migration considerations for velocity
Migration considerations for discretionary goal management
Migration considerations for dynamic alias management
Migration considerations for multisystem enclaves
Migration considerations for protection of critical work
Migration considerations for managing non-enclave work in enclave servers
Migration considerations for an increased notepad size
WLM managed batch initiator balancing
Consider resource group maximum in WLM batch initiator management
Defining a coupling facility structure for multisystem enclave support
Defining the coupling facility
Shutting down the coupling facility
Coupling facility failures
The Intelligent Resource Director
LPAR CPU management
Dynamic channel path management
Channel subsystem priority queuing
Example: How the Intelligent Resource Director works
Making the Intelligent Resource Director work
Defining the SYSZWLM_xxxxyyyy coupling facility structure
Enabling LPAR CPU management
Enabling non-z/OS CPU management
Enabling dynamic channel path management
Enabling channel subsystem priority queuing
For more information
Using System z Application Assist Processor (zAAP)
Performing capacity planning to project how many zAAPs will be needed (zAAP Projection Tool)
Meeting software and hardware requirements associated with the zAAPs
Acquiring the zAAPs
Defining zAAPs to the desired LPARs
Reviewing parameter settings associated with zAAP usage
Review z/OS parameter settings
Review Java parameter settings
Considering automation changes related to zAAP usage
Monitoring zAAP utilization and configuring changes appropriately
Using System z Integrated Information Processor (zIIP)
Meeting software and hardware requirements for using zIIPs
Planning for zIIPs
Acquiring zIIPs
Defining zIIPs
Reviewing z/OS parameter settings
Using zIIPs — miscellaneous services
Activating zIIPs
Using the WLM ISPF application
Before you begin
Panel areas and how to use them
Using the menu bar
Using the menu bar on selection lists
Using the status line
Using the scrollable area
Using the menu bar on a selection list
Using the Action field
Using the command line
Using the function keys
Starting the WLM application
Now you're started
Using the Definition Menu
Using the menu bar on the Definition Menu
File
Utilities
Notes
Options
Help
Working with service policies
Working with workloads
Working with resource groups
Working with service classes
Defining goals
Using action codes on service class panels
Defining service policy overrides
Working with classification rules
Using action codes on the Modify Rules panel
Using selection lists for classification rules
Creating a subsystem type for rules
Deleting a subsystem type for rules
Working with classification groups
Working with report classes
Working with service coefficients and options
Working with application environments
Working with scheduling environments
Creating a new scheduling environment
Modifying a scheduling environment
Copying a scheduling environment
Browsing a scheduling environment
Printing a scheduling environment
Deleting a scheduling environment
Creating a new resource
Creating a new resource directly from the main panel
Creating a new resource while working with a scheduling environment
Summary of the two methods for creating new resources
Showing all cross-references for a resource definition
Deleting a resource
Working with the guest platform management provider (GPMP) settings
Coordinating updates to a service definition
Using the WLM couple data set
Allocating a WLM couple data set
Installing and extracting a service definition
Using MVS data sets
Restricting access to your service definition
Activating a service policy
Printing in the application
Browsing definitions
Using XREF function to view service definition relationships
WLM application messages
IWMAM040
IWMAM041
IWMAM042
IWMAM043
IWMAM044
IWMAM046
IWMAM047
IWMAM050
IWMAM051
IWMAM052
IWMAM054
IWMAM055
IWMAM058
IWMAM072
IWMAM077
IWMAM098
IWMAM099
IWMAM313
Using the z/OS Management Facility (z/OSMF) to administer WLM
Overview of the z/OSMF workload management task
Key functions of the Workload Management task in z/OSMF
Customizing the WLM ISPF application
Specifying the exits
Coding the WLM exits
IWMARIN1
Customizing the WLM application libraries — IWMAREX1
Customizing the WLM application data sets — IWMAREX2
Adding WLM as an ISPF menu option
Moving pop-up windows
Customizing the keylists
CPU capacity table
Using SMF task time
Return codes for the IWMINSTL sample job