Current and future sysplex considerations

To ensure the best configuration of Total Object Manger (TOM) for your enterprise, consider the following factors when configuring the product:

  • Using one TOMPLEX versus separate TOMPLEXs
  • Number of systems or sysplexes
  • Disaster recovery procedures and facilities
  • Sysplex (multi-TOM TOMPLEX) versus monoplex (single-TOM TOMPLEX)
  • Operations staff skill level

Related topics

TOM-best-practices

Planning-and-testing

To make the appropriate decisions that result in the best configuration of the operation and administration of the objects, applications, and workloads on your systems and sysplexes, review the following questions:

Considerations

Questions

Using one TOMPLEX versus separate TOMPLEXs

  • Do you require one TOMPLEX or separate TOMPLEXs for your development, testing and production systems?
  • How do you roll out applications?
  • How do you handle change management?

Number of systems or sysplexes

  • How many systems or sysplexes do you have?
  • Do you want TOM to manage JES2?
  • Do you want to manage only started tasks or do you want to manage other resources, such as USS processes, WLM resources, and other generic resources?

Sysplex (multi-TOM TOMPLEX) versus monoplex (single-TOM TOMPLEX)

  • Do you use moveable workloads, objects, or applications within the sysplex that require you to use a multi-TOM TOMPLEX?
  • How do you want to administer your definition bases? 
    • Are automation personnel roles distributed among many departments?
    • Do you have project groups, DBAs, and so on?
    • Who owns the objects?
    • How do you manage the process or change management control?
    • Does your sysplex have sufficient shared DASD on which the registry VLDS can be allocated for sharing among the TOMs on the systems in a multi-TOM TOMPLEX?
  • Do you use separate security databases or policies among your systems or sysplexes?

Operations staff skill level

  • (Recommended) Do you train your staff to use operator commands and dialog boxes such as STOP, SHUTSYS, START, LOCK, and RESET?

  • Do you maintain custom automation (EXECs and BMC AMI OpsA Rules) so that your operations staff doesn't have to learn anything new?

    Important

    For example, you can define CMD-type Rules to intercept the MVS START and STOP commands and convert them to TOM START and STOP commands.

    However, this causes the following issues:

    • Requires strict naming conventions (Rule must associate the object name with the jobname)
    • Might require enabling TOM automation variables (overhead)
    • Disables the MVS START and STOP commands

Workload operations

  • When should a resource be active?
    • Do you know if and how a resource relates to others running in the enterprise?
    • Must one resource be active for another resource to become active?    
    • On what systems can the resources be active?
    • Does a resource act as a server to others? 
  • Do you need a schedule?
    • Are there holidays that might alter the schedule?

Disaster recovery procedures and facilities

  • How do you handle disaster recovery?
  • Who gets notified?
  • Must damage control rely on the feedback of a person or can TOM make the appraisal and initiate corrective action?

Implementing TOM

  • Did you implement the TOM stated task and other supporting address spaces such as the CAS, the BMC AMI Ops Automation PAS, and MainView Explorer?
  • Have you properly defined and identified your UNIX file system structure to the CAS?
  • Are you migrating from another availability manager and have custom automation that relies on automation items that you need to modify for TOM?
  • What do you want TOM to manage?

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*