UHH>Informatik>NatS>User Web>TWikiUsers>KimberlyKilman>MaxmChapter (17 Oct 2012, UnknownUser) Print version

Maxm Chapter

-- KimberlyKilman -- 17 Dec 2004
  1. Classic MAXM Reorg Solutions
    for IMS

MAXM Reorg for IMS and MAXM Reorg for IMS with Online/Defrag solutions include the classic database utilities.

  • Reorg function
  • Unload function
  • Reload function
  • Index Build function
  • Prefix Resolution and Update function
  • Online/Defrag function (optional)

With the classic MAXM Reorg solutions, a database, including all indexes, can be reorganized in one step. One set of JCL runs the entire reorganization process. To learn more about the features and benefits of each function, refer to

All MAXM solutions include the DBA Toolkit. For information about the toolkit, refer to

System Flow

The Reorg function manages reorganization (unload, reload, and index builds) in a single-step. For databases that contain logical relationships, the Prefix Resolution and Update function runs in a separate job step. All functions can run as stand-alone utilities.

When the Reorg function manages a reorganization, input and output databases use different data sets. A segment is unloaded and passed directly to the Reload function. Real storage usage does not increase because segments are passed directly and not stored in hyperspace.

The Reorg function passes WF1 records from the Reload function directly to the Index Build function. The process continues for each segment. The Index Build function sorts the index records and creates index databases.

The Reorg function does not attempt to manage all reorganization tasks for databases that contain logical relationships. Additional processing is required after the reorganization to resolve logical relationships.

DBRC control is essential to the integrity and success of the reorganization process. BMC Software recommends using DBRC with MAXM Reorg.

See MAXM Reorg System Flow shows the MAXM Reorg system flow.

 
 
 
 

The Online/Defrag function executes independently from all other MAXM Reorg functions. For more information about the Online/Defrag function, refer to

 
 
MAXM Reorg System Flow
 

Reorganization Overview

Use the MAXM Reorg functions to reorganize a database. Using the Reorg function to coordinate all functions simplifies reorganization. However, you can use the functions as stand-alone utilities.

Reorg Function

This section provides an overview of the steps that are involved when you use the Reorg function to reorganize a database.

 
 
 
 

Use the JCL generator to generate optimized JCL and your reorganization process is simple.
 
 
  1. Required. Install MAXM Reorg. For instructions, see the IMS Database Utilities Installation Guide .
  2. Required. Prepare the databases for reorganization. Each database must be offline or in read only mode.

Issue a /STA DB dbdname ACCESS=RD command for each database to change the DBRC access intent.

Issue one of the following commands for each database that is to be reorganized:

  • If you specify SHARE(YES) to allow read access during the reorganization, issue a /DBD dbdname NOFEOV command for each database.
  • If you specify SHARE(NO) , issue a /DBR dbdname NOFEOV command for each database.

Issue the /DBD or /DBR commands in a separate step before the reorganization job step. If you specify AUTHERR(WTOR) in the Reorg function job step and you have not issued the /DBD or /DBR commands for all databases, the Reorg function issues messages stating which databases must be stopped.

  1. Required. Create and submit one set of JCL that runs the entire reorganization process. You can use your existing JCL for the Reload function and add a few statements specific to the Reorg function. You also need to add IDCAMS statements to allocate the output data sets.
  2. Optional. If you want any special processing, set parameters in the PLUSIN DD statement.
  3. Optional. If you have logical relationships, invoke the Prefix Resolution and Update function.
  4. Required. Swap the data set names. If you set SWAP(Y) in the PLUSIN data set and you did not reorganize databases with logical relationships, the Reorg function automatically swaps the data set names.

Stand-Alone Utilities

This section provides an overview of the steps that are involved when you use MAXM Reorg functions as stand-alone utilities to reorganize a database.

 
 
 
 

When using stand-alone utilities, you must create and submit separate sets of JCL for each job step.
 
 
  1. Required. Install MAXM Reorg. For instructions, see the IMS Database Utilities Installation Guide .
  2. Required. Take the database offline.
  3. Required. Take an image copy of the database before reorganization. If an error occurs during any reorganization step, you can recover to this image copy.
  4. Required. Unload the database with the Unload function.
  5. Required. Delete and define any database data sets that were not defined as reusable so that the data can be reloaded.
  6. Required. Reload the database with the Reload function.
  7. Required. Rebuild the indexes with the Index Build function.
  8. Required. Resolve and update prefixes for databases involved in logical relationships with the Prefix Resolution and Update function.
  9. Required. Take an image copy of the reorganized database.
  10. Required. Start the database.

System Requirements

MAXM Reorg has the following system requirements:

Hardware MAXM Reorg operates on any CPU that supports any IBM-supported version of MVS. The hardware requirements are the same as those for the IBM IMS product. MAXM Reorg is written in assembler language within the constraints of the assembler for OS/VS2.

MVS MAXM Reorg supports all IBM-supported versions of MVS and will support new versions as they become available from IBM.

IMS You must be a licensed user of the IMS/ESA Database Manager product. MAXM Reorg supports all IBM-supported versions of IMS and will support new versions as they become available from IBM.

Sort Product The Unload function and Index Build function can use IMS DL/I and a sort product, depending on the function being performed. The sort product must be compatible with the IBM OS/VS Sort/Merge Program. The Prefix Resolution and Update function requires an OS/VS sort/merge utility that is compatible with the IBM DFSORT product.

ISPF To use the optional ISPF interface, IBM Interactive System Productivity Facility (ISPF) version 2.3 (or later) is required.

DBUSS The DBUSS (required for the Unload function; optional for the Index Build function) has the following system requirements:

  • IBM-supported version of MVS
  • APF-authorized library
  • CSA requirements:
  • 15 KB ECSA/CSA for the subsystem base
  • 1,040 bytes CSA (allocated below the line)
  • 7 KB ECSA/CSA for the Unload function support in the subsystem
  • 552 bytes CSA for each data set group being unloaded by an Unload function job using the subsystem. Although 552 bytes (plus 552 bytes for each additional data set group) is required when using the subsystem, it is allocated to the Unload function rather than the subsystem.

CICS Local DL/I is not supported. For more information, see the IMS Database Utilities Installation Guide .

Installation and Maintenance
MAXM Reorg is easy to install. The Quick Install System (QIS) provides an easy-to-use, interactive method for creating installation JCL. The manual installation procedure provides the same results as the DBU Install System, but you must tailor the JCL manually.

Because BMC Software provides complete replacements for all product modules, the procedure for installing maintenance is similar to the initial installation procedure. Installing product maintenance does not require a new CPU ID password.

For more information, see the MAXM Reorg for IMS Solutions Installation Guide for more information.

Limitations

This section describes limitations of MAXM Reorg.

 
 
 
 

MAXM Reorg does not support the IBM IMS/ESA Year 2000 Exit Tool product.
 
 

Reorg Function

The Reorg function has the following limitations:

  • Certain DBD characteristics cannot be changed between unload and load processes.
  • The Reorg function cannot be used to rebuild secondary indexes that are shared between different primary databases.
  • The Reorg function does not accept all keywords from other MAXM Reorg functions.
  • All limitations of the Unload function, Reload function, and Index Build function are limitations of the Reorg function.

Unload Function

The Unload function has the following limitations:

  • ISAM/OSAM for any database organization are not supported.
  • Parallel unloading of the database when root keys are compressed, regardless of database organization, is not supported. The Unload function does support HDAM root segments with compressed keys
    for all other functions and features.
  • HSAM and SHSAM databases are not supported.
  • In an HDAM database, all root segments must have a sequence (key) field. The sequence field does not have to be unique.
  • The DB Scan function cannot run under Utility Control Facility (UCF).
  • The DB Scan function does not support partitioned databases.
  • When reading a multi-volume OSAM database without the DBUSS, you must first run the BMC Software-provided program DBUVTOC. For more information, see the IMS Database Supplemental Utilities Reference Manual for more information.

Reload Function

The Reload function has the following limitations:

  • For HDAM DBDs, you must specify a maximum relative block number (RBN) in the RMNAME operand of the DBD macro.
  • All root segments in an HDAM database must have a sequence (key) field. The sequence field does not have to be unique.
  • The Reload function does not provide direct support for HISAM unload and reload utilities. The Reload function can reload the HD Unload data set of a HISAM database. The Reload function does not support HISAM databases with multiple data set groups. This limitation applies to HISAM ISAM/OSAM databases only.
  • The Reload function allocates multi-volume OSAM data sets with a single DD statement. If you allocate a multi-volume OSAM data set with multiple DD statements, you must execute the BMC Software-provided DBUVTOC program before using the Reload function. For more information, see the IMS Database Supplemental Utilities Reference Manual .
  • If you are running the Reload function under the UCF, the Reload function user exit facility is not supported. The Reload function does support the UCF user exit facility.
  • The Reload function does not support the use of the Load Program Interface (LPI) with the UCF.

Index Build Function

The Index Build function has the following limitations:

  • The HISAM Reorganization Reload utility is required to load a partial secondary index database.
  • If you are using the Unload function to scan a pre-allocated multi-volume OSAM database and you are not using the DBUSS for fast I/O, you must have executed the BMC Software-provided program DBUVTOC after the allocation. For more information, see the IMS Database Supplemental Utilities Reference Manual.
  • When using the IDCAMS external program function (deleting/defining VSAM data sets within the PLUSIN data set), you cannot delete a cluster from one volume and allocate it to a different volume within the same Index Build function step.
  • If you use the IBM Storage Management System (SMS), you also must delete or define your VSAM clusters in a separate step before the Index Build function step.
  • If the target segment in the primary database is a non-root segment and it is not the target of an existing secondary index, you must create a physical parent pointer by running a complete reorganization. Use the old DBD for the unload step; use the new DBD, with the new secondary index defined, for the reload and secondary index creation steps.

Prefix Resolution and Update Function

The Prefix Resolution and Update function has the following limitations:

  • Only execute the IMS region controller with the following programs:
  • user database load programs
  • BMC Software or IMS database reorganization reload utilities
  • Prefix Update (unless you are using PRPURGP0 or the Concurrent Prefix Update function)
  • The function cannot run the Prefix Resolution function under the region controller.
  • The function cannot use the Work File Generator when building multiple databases in one job step.
  • Because the Prefix Resolution and Update function does not use the IMS buffer handler to update databases, logging is not supported. No IEFRDER DD statement is required.
  • The Prefix Update function cannot process the data set created by the IMS Prefix Resolution utility because the relative byte addresses (RBAs) are not always in physical sequence.
  • When less than three percent of the total number of database records are updated during the prefix update process, the IMS Prefix Update utility may perform better than the Prefix Update function. For better performance, separate the database data sets that have very few updates from all other data set groups and process them with the IMS Prefix Update utility.
  • The Prefix Resolution and Update function does not support HISAM databases. For HIDAM databases that contain logical relationships, split the output of the Logical Twin Resolution phase and run the IMS Prefix Update utility.
  • The Prefix Resolution and Update function cannot execute under the UCF.
  • The Prefix Update function does not update logical parent (LP) counter fields to zero. These fields are updated only when you delete a segment type involved in a logical relationship. In this case, specify FORCELP(Y) in the PLUSIN control statement data set during Prefix Resolution. FORCELP(Y) is available only through the PLUSIN control statement.
  • When running the Prefix Update function for a multi-volume OSAM database, you must first run the BMC Software-provided program DBUVTOC. For information on this program, see the IMS Database Supplemental Utilities Reference Manual .

Online/Defrag Function

The Online/Defrag function has the following limitations:

  • HDAM and HIDAM databases are supported. Limited HALDB support. No other database types are supported.
  • Logical databases are not supported.

 

 
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback