Further Reading — Chapter 29

IBM Documentation

RMF and SMF

  • z/OS Resource Measurement Facility (RMF) User's Guide (SC34-2664) — The definitive reference for RMF configuration, report interpretation, and performance data collection. The chapters on Monitor I (long-term reporting) and the Post-Processor are essential reading for capacity planners. Pay particular attention to the duration report format and the workload activity report.
  • z/OS MVS System Management Facilities (SMF) (SA38-0667) — SMF record type reference. For capacity planning, focus on Type 70 (processor activity), Type 72 (workload activity), Type 74 (device activity), and Type 89 (sub-capacity data). The self-defining section format described in Chapter 27 applies to these records as well.
  • z/OS RMF Performance Management Guide (SC34-2665) — Goes beyond the mechanics of RMF to discuss performance management methodology. The chapter on capacity planning methodology is directly relevant to Section 29.7 of this chapter.

WLM and Capacity

  • z/OS MVS Planning: Workload Management (SY38-0686) — Revisit Chapter 5's WLM concepts with specific attention to service class definitions and their relationship to capacity measurement. The section on WLM reporting and SMF Type 72 data collection connects directly to the workload characterization approach described in this chapter.
  • z/OS Workload Management Services Guide and Reference — API documentation for programmatic WLM data access, useful for building custom capacity analysis programs.

Pricing and Licensing

  • IBM Z Software Pricing Reference — IBM's official documentation of MLC and sub-capacity pricing mechanics. Updated periodically. Understanding the SCRT calculation and R4HA mechanics described in Section 29.5 requires familiarity with this document.
  • IBM Tailored Fit Pricing for z/OS — IBM's documentation of TFP/ECS and Enterprise Capacity Solution. Critical for the MLC vs. TFP analysis described in Section 29.5.1.
  • Sub-Capacity Reporting Tool (SCRT) User's Guide — How SCRT calculates billable MSUs from SMF data. Every capacity planner should understand this calculation — errors in SCRT reporting have direct financial consequences.

Hardware and Configuration

  • IBM z16 Technical Introduction (SG24-8460) — Hardware specifications, processor ratings, and capacity characteristics. The LSPR (Large Systems Performance Reference) data referenced in this chapter is published alongside technical introductions for each hardware generation.
  • z/OS Capacity Provisioning User's Guide (SC34-2661) — IBM's Capacity Provisioning Manager automates CoD activation based on WLM workload analysis. Directly relevant to the R4HA management techniques in Section 29.5.3.

Vendor Documentation

Capacity Planning Tools

  • BMC Capacity Management for z/OS — Comprehensive capacity planning product that automates data collection, trend analysis, and forecasting. Covers historical data management, what-if modeling, and reporting. Relevant to the data pipeline described in Section 29.2.4.
  • Broadcom (CA) Capacity Management — Capacity planning and optimization suite covering performance data collection, trend forecasting, and cost analysis. The chargeback module is relevant to the MSU budgeting discussion.
  • IBM Z Capacity Analytics (zCA) — IBM's analytics platform for z/OS capacity planning. Uses machine learning for forecasting and anomaly detection. Provides the automated trend analysis and exception detection described in Section 29.7.

Performance Monitoring (Capacity Planning Data Sources)

  • BMC MainView for z/OS — Real-time and historical performance monitoring. The historical data repository feeds capacity planning analysis.
  • Broadcom SYSVIEW Performance Management — Performance monitoring with historical trend analysis. Useful for the capacity data pipeline described in this chapter.
  • IBM OMEGAMON for z/OS — Real-time and historical monitoring. The OMEGAMON data warehouse feature supports long-term capacity trend analysis.

Books

  • z/OS Capacity Planning and Performance Management (IBM Redbook, SG24-8047) — The single most relevant IBM publication for this chapter. Covers the entire capacity planning lifecycle: data collection, workload characterization, forecasting, and optimization. The chapters on MSU budgeting and SCRT analysis are particularly valuable.
  • z/OS Performance: Monitoring and Tuning (SG24-6192) — Companion to the capacity planning Redbook. Focuses on the technical measurement and analysis techniques that feed capacity planning. The RMF sections are essential.
  • ABCs of z/OS System Programming, Volume 10: Capacity Planning (SG24-6990) — Accessible introduction to capacity planning concepts for those who need a foundation before tackling the advanced material in this chapter.
  • Large Systems Performance Reference (LSPR) — IBM's processor benchmark data, published for each hardware generation. This is the source for MSU/MIPS ratings and cross-generation performance comparisons. Available through IBM's Z systems performance website.
  • IT Financial Management by Maxime Sottini (Van Haren Publishing, 2012) — Covers IT financial management principles including capacity-based cost allocation, chargeback models, and IT budget construction. Useful context for the MSU budgeting discussion in Section 29.5.
  • The Art of Capacity Planning by John Allspaw (O'Reilly, 2008) — While focused on web-scale systems, the principles of capacity planning, forecasting methodology, and organizational process apply directly to mainframe environments. The chapter on "When Capacity Planning Fails" provides cautionary examples that parallel the stories in Section 29.1.
  • Capacity Planning for Web Services: Metrics, Models, and Methods by Daniel A. Menasce and Virgilio A.F. Almeida (Prentice Hall, 2001) — Academic treatment of capacity planning methodology, including queuing theory and statistical forecasting models. The mathematical foundations for the regression and seasonal adjustment techniques in Section 29.4.
  • Site Reliability Engineering: How Google Runs Production Systems by Betsy Beyer et al. (O'Reilly, 2016) — The chapter on capacity planning at Google provides a modern perspective on the same principles applied in mainframe capacity planning. The concept of "provisioning for peak + headroom" translates directly to the 75th-percentile planning approach described in this chapter.

Technical Articles and Papers

  • "Understanding MSU and Software Pricing on IBM Z" — IBM Z Content Solutions publishes periodic articles explaining MSU calculation, SCRT mechanics, and TFP pricing models. Essential reading for anyone involved in mainframe cost management.
  • "zIIP Offload Opportunities for COBOL Workloads" — Search IBM Z and mainframe community sites for articles on maximizing zIIP offload. The DRDA and z/OS Connect offload techniques described in Section 29.2.2 are covered in greater detail.
  • "Capacity Planning in a Hybrid Cloud Environment" — IBM and analyst firms (Gartner, Forrester) publish whitepapers on capacity planning that spans mainframe and cloud environments. Relevant to the modernization capacity impact analysis in Case Study 2.
  • SHARE Conference Proceedings — The SHARE user group conference regularly features sessions on capacity planning, MSU optimization, and pricing strategy. Search the proceedings archive for "capacity planning," "MSU management," and "software pricing."
  • CMG (Computer Measurement Group) Publications — CMG is the professional organization for performance and capacity management practitioners. Their conference proceedings and technical journal contain advanced capacity planning methodologies, statistical forecasting techniques, and vendor-neutral best practices.

Online Resources

  • IBM Large Systems Performance Reference (LSPR) (ibm.com) — Official processor benchmark data used for MSU/MIPS ratings. Updated with each new hardware generation. Bookmark this — you'll reference it every time you do a hardware sizing exercise.
  • IBM Z and LinuxONE Community (community.ibm.com) — Active forums where practitioners discuss capacity planning strategies, SCRT analysis, and pricing optimization. The "Performance and Capacity" forum is directly relevant.
  • Planet Mainframe (planetmainframe.com) — Industry articles on mainframe capacity management, pricing trends, and optimization strategies.
  • Mainframe Analytics (mainframeanalytics.com) — Independent analysis of IBM Z pricing models, MSU optimization techniques, and capacity planning best practices.
  • CMG International (cmg.org) — Professional organization for performance and capacity management. Membership provides access to a library of capacity planning papers, case studies, and methodologies.
  • Chapter 1 (z/OS Ecosystem) — Hard prerequisite. The z/OS architecture, subsystem structure, and LPAR concepts are foundational to capacity planning.
  • Chapter 5 (Workload Manager) — Hard prerequisite. WLM service classes determine workload characterization for capacity planning. WLM capping is a capacity optimization lever. SMF Type 72 (WLM reporting) is a primary data source.
  • Chapter 6 (DB2 Optimizer) — Soft prerequisite. DB2 access path decisions drive CPU consumption. Capacity impact analysis for DB2 changes requires understanding optimizer behavior.
  • Chapter 23 (Batch Window Engineering) — Soft prerequisite. The batch window represents peak capacity consumption at most shops. Batch scheduling optimization is an R4HA management technique.
  • Chapter 26 (Batch Performance) — Soft prerequisite. Batch performance optimization directly reduces capacity requirements. The performance decomposition (CPU, I/O, DB2, other) applies to capacity impact analysis.
  • Chapter 27 (Batch Monitoring) — The SMF data pipeline and monitoring infrastructure from Chapter 27 feeds capacity planning. Exception reporting uses the same alerting framework.
  • Chapter 28 (Mainframe Security) — Security features (encryption, RACF, audit logging) consume capacity that must be budgeted. The security overhead estimates from Chapter 28 are inputs to the capacity plan.
  • Chapter 30 (Disaster Recovery) — Forward reference. The capacity plan determines DR site sizing requirements. The DR site must have sufficient capacity to run the recovery workload.
  • Chapter 31 (Operational Automation) — Forward reference. Automated capacity management (CoD activation, workload balancing) reduces manual intervention during peak periods.