Front cover -- Contents -- Figures -- Tables -- Notices -- Trademarks -- Preface -- The team that wrote this redbook -- Become a published author -- Comments welcome -- Chapter 1. VSAM basics -- 1.1 A brief description of VSAM -- 1.2 VSAM functions by release level -- 1.3 What is VSAM? -- 1.3.1 VSAM access types -- 1.4 Major VSAM parts -- 1.4.1 Catalog management -- 1.4.2 Record management -- 1.5 VSAM terminology and concepts -- 1.5.1 Logical record -- 1.5.2 Key field -- 1.5.3 Ways to identify logical records -- 1.5.4 Physical record -- 1.5.5 Control interval -- 1.5.6 Control area -- 1.5.7 Spanned records -- 1.5.8 Component -- 1.5.9 Cluster -- 1.5.10 Sphere -- 1.5.11 Alternate indexes -- 1.5.12 Splits -- 1.5.13 VSAM buffering -- 1.6 VSAM data set organizations -- 1.6.1 Key-sequenced data set -- 1.6.2 Entry sequenced data set (ESDS) -- 1.6.3 Relative record data set -- 1.6.4 Variable relative record data set -- 1.6.5 Linear data set (LDS) -- 1.7 Comparing VSAM data set organizations -- 1.8 Choosing a VSAM data set type -- 1.9 Extended format data set -- 1.10 Extended addressability -- 1.11 Data striping -- 1.12 Processing a VSAM cluster -- 1.12.1 Allocating a VSAM cluster -- 1.12.2 Accessing VSAM cluster -- 1.12.3 Unallocation -- 1.13 VSAM exploiters -- 1.13.1 DB2 -- 1.13.2 Hierarchical file system (HFS) -- 1.13.3 zSeries File System (zFS) -- 1.13.4 CICS -- 1.13.5 DFSMShsm -- 1.13.6 DFSMSrmm -- 1.13.7 Java Record I/O (JRIO) -- Chapter 2. Performance -- 2.1 Service level agreement -- 2.2 Transaction performance -- 2.3 Performance management -- 2.3.1 I/O performance -- 2.4 VSAM performance management -- 2.5 VSAM rule-of-thumb mode -- 2.5.1 Invalid rules-of-thumb -- 2.6 Parameters affecting performance -- 2.6.1 Allocation units -- 2.6.2 Guaranteed Space -- 2.6.3 Optimizing control area (CA) size -- 2.6.4 Partial release.

2.6.5 Allocation constraint relief -- 2.6.6 Control interval size -- 2.6.7 FREESPACE definition for KSDS and ESDS -- 2.6.8 Index options -- 2.6.9 Key Range and Ordered -- 2.6.10 Share options -- 2.6.11 Initial load option -- 2.6.12 Region size -- 2.6.13 Buffering options -- 2.6.14 Buffering techniques -- 2.6.15 Data compression -- 2.6.16 VSAM Data striping -- 2.7 VSAM performance by scenarios -- 2.7.1 Performance scenario using RMF reports -- 2.7.2 Reducing the number of I/Os -- 2.7.3 I/O wait time (IOSQ) for VSAM data sets -- 2.7.4 I/O service time (connect) for VSAM data sets -- 2.7.5 Decreasing VSAM CPU time -- 2.8 VSAM and ESS controllers -- 2.8.1 ESS model 800 enhancements -- 2.8.2 Lab experiments -- 2.9 Performance monitors -- 2.9.1 Resource measurement facility -- 2.9.2 Tivoli Decision Support (TDS) -- 2.9.3 Generalized Trace Facility (GTF) -- Chapter 3. VSAM problem determination and recovery -- 3.1 VSAM problem determination hints and tips -- 3.1.1 How to check your VSAM data set -- 3.1.2 z/OS system messages -- 3.1.3 Catalog Search Interface IGGCSIVS program -- 3.1.4 System LOGREC messages -- 3.1.5 GTF CCW traces -- 3.1.6 DITTO/ESA output -- 3.1.7 What can you get from the SMF records? -- 3.2 Some common VSAM problems -- 3.2.1 Lack of virtual storage -- 3.2.2 Initial loading problems -- 3.2.3 Mismatch between catalog and data set -- 3.2.4 Hardware errors -- 3.2.5 Bad data or bad channel program -- 3.2.6 Structural damage -- 3.2.7 Improper sharing -- 3.2.8 Mismatch between catalog and VTOC -- 3.2.9 VSAM does not produce expected output -- 3.2.10 VSAM RLS problems -- 3.2.11 VSAM and DFSMStvs considerations -- 3.2.12 OEM problems -- 3.2.13 Enqueue issues -- 3.2.14 Migration issues -- 3.2.15 Performance considerations -- 3.2.16 Deadlocks -- 3.2.17 Beware of some VSAM restrictions -- 3.3 What documentation to collect.

3.3.1 Catalog performance problems -- 3.3.2 VSAM RLS problems -- 3.3.3 IDCAMS problems -- 3.3.4 Broken VSAM data set -- 3.3.5 Broken catalog -- 3.3.6 How to obtain VSAM record management trace? -- 3.4 How to recover a damaged VSAM data set -- 3.4.1 EXAMINE command -- 3.4.2 DIAGNOSE command -- 3.4.3 VERIFY command -- 3.4.4 Broken Index scenario -- 3.4.5 Abend task scenario -- 3.4.6 Recovering damaged BCS entries -- 3.4.7 Recovering damaged VVDS entries -- 3.5 Prevention is better than cure -- 3.5.1 Back up your VSAM data sets -- 3.5.2 Keep your system at current maintenance levels -- 3.5.3 Use Resource Recovery Management Services (RRMS) -- 3.6 Where to look for more information -- 3.6.1 IBM manuals and sources of relevant information -- 3.6.2 Information APARs from IBMLINK on VSAM problems -- 3.6.3 Information APARs on specific problems -- 3.6.4 VSAM information on the Internet -- 3.7 IDC3009I message -- 3.8 IDCAMS LISTCAT output fields -- 3.8.1 High used RBA value (HURBA) for KSDS -- 3.8.2 High allocated RBA value (HARBA) -- 3.8.3 FREESPC -- 3.8.4 High key RBA/CI -- 3.8.5 High-level index RBA value -- 3.8.6 Sequence set first RBA value -- 3.8.7 Number of index levels -- 3.8.8 Time stamps -- 3.9 SMF record types related to VSAM data sets -- 3.9.1 SMF record type 60 -- 3.9.2 SMF record type 61 -- 3.9.3 SMF record type 62 -- 3.9.4 SMF record type 63 -- 3.9.5 SMF record type 64 -- 3.9.6 SMF record type 65 -- 3.9.7 SMF record type 66 -- 3.9.8 SMF record type 67 -- 3.9.9 SMF record type 68 -- 3.9.10 SMF record type 69 -- 3.9.11 SMF record type 42 -- 3.10 RRMS and VSAM -- Chapter 4. Managing your VSAM data sets -- 4.1 Reorganization considerations -- 4.1.1 CI/CA splits -- 4.1.2 The loss of useful space in Data CA -- 4.1.3 CI/CA splits causing free space increase -- 4.2 New Index CI size calculation algorithm -- 4.2.1 Analyze existing data sets.

4.3 Sharing VSAM data sets -- 4.3.1 Write and read integrity -- 4.3.2 VSAM sharing mechanisms -- 4.3.3 Sharing data in a single VSAM control block structure -- 4.3.4 Sharing data with many VSAM control block structures -- 4.3.5 General share options: Considerations -- 4.3.6 Protecting VSAM data set through DISP parameter -- 4.4 Extended addressability (EA) -- 4.5 Catalog Search Interface -- 4.5.1 CSI setup -- 4.6 Major sources of VSAM processing options -- 4.6.1 ACB control block -- 4.6.2 DD statement keywords -- 4.6.3 Catalog BCS and VVDS entries -- 4.6.4 SMS constructs -- 4.7 Media Manager, Open, Close, EOV in VSAM -- 4.7.1 OPEN macro -- 4.7.2 CLOSE macro -- 4.7.3 End-of-Volume (EOV) -- 4.8 VSAM and 64 bits -- 4.9 Special considerations for COBOL users and SMB -- 4.9.1 COBOL users take note -- Chapter 5. VSAM Record Level Sharing -- 5.1 Introducing VSAM RLS -- 5.1.1 What is VSAM RLS? -- 5.1.2 Why RLS? -- 5.1.3 How does RLS work? -- 5.1.4 RLS in a single system (monoplex) -- 5.1.5 CICS and VSAM RLS -- 5.1.6 RLS restrictions -- 5.2 RLS terminology -- 5.3 Planning for RLS -- 5.3.1 Hardware requirements -- 5.4 Implementing VSAM RLS -- 5.4.1 Define Sharing Control Data Set (SHCDS) -- 5.4.2 Define CF cache structures -- 5.4.3 Define CF lock structures -- 5.4.4 SMS definitions -- 5.4.5 Modifying the PARMLIB IGDSMSxx Member -- 5.4.6 Security definitions -- 5.4.7 Using a VSAM sphere in RLS mode -- 5.4.8 RLS Recoverable spheres -- 5.5 RLS problem determination and recovery -- 5.5.1 Problems with SHCDS -- 5.5.2 Problems with SMSVSAM -- 5.5.3 Problems with locks -- 5.5.4 SHCDS FALLBACK procedure -- 5.5.5 RLS rules -- 5.5.6 MVS commands for RLS -- 5.6 RLS enhancements -- 5.6.1 RLS/KSDS extended addressability -- 5.6.2 VSAM RLS CF structure duplexing and rebuild -- 5.6.3 RLS CF caching enhancements -- 5.7 RLS performance.

5.7.1 Factors affecting RLS data sharing performance -- 5.7.2 CF access time -- 5.7.3 RLS performance gains -- 5.7.4 CICSPlex RLS performance comparison -- 5.7.5 Batch RLS performance experiments and comparison -- 5.7.6 RMF and VSAM RLS -- 5.7.7 MVS commands about RLS performance -- 5.7.8 SMF records covering VSAM RLS -- Chapter 6. DFSMStvs -- 6.1 Introducing DFSMStvs -- 6.2 Why DFSMStvs? -- 6.2.1 How to extend CICS availability -- 6.2.2 Reducing the batch window -- 6.3 Some definitions -- 6.3.1 Backward recovery -- 6.3.2 Forward recovery -- 6.3.3 Atomic updates -- 6.3.4 Unit of work and unit of recovery -- 6.3.5 Two-phase commit -- 6.3.6 In-flight and in-doubt -- 6.3.7 Repeatable read -- 6.3.8 Recoverable data sets -- 6.4 CICS support for recoverable VSAM -- 6.5 DFSMStvs overview -- 6.5.1 The RLS connection -- 6.5.2 DFSMStvs locking -- 6.5.3 DFSMStvs logging -- 6.5.4 Recovery coordination -- 6.6 Our experiences with implementation -- 6.6.1 Define list structures in the CFRM policy -- 6.6.2 Define the log structures and log streams in LOGR policy -- 6.6.3 Define SMS constructs for DFSMStvs -- 6.7 DFSMStvs problem determination tips -- 6.7.1 How to take a dump of the problem? -- 6.7.2 Classes of errors -- 6.7.3 Determining the Failing Module -- 6.7.4 Apparent batch job hangs -- 6.7.5 Other hangs -- 6.7.6 Quiescing a data set -- 6.7.7 Close/delete/rename of data set with inflight UR -- 6.7.8 New and changed system level commands for DFSMStvs -- 6.7.9 SET SMS and SETSMS commands -- 6.7.10 VARY SMS command -- 6.7.11 SYS1.PARMLIB changes -- 6.7.12 Changes to Job Control Language (JCL) -- 6.7.13 Changes to IDCAMS -- 6.7.14 Messages and codes -- 6.7.15 Macros that have been changed to support DFSMStvs -- Appendix A. Sample code -- JRIO API examples -- Locate a record by key in keyed access record file -- Position to a record in a random access record file.

Read a record from a keyed access record file.