DCMS Disk Management Suite - DMS
|
Disk Management Suite - DMS |
|
DMS is a comprehensive collection of tools that help you manage your mainframe disk storage environment. It is an invaluable add-on to your IBM HSM implementation. The DMS utilities fall into 6 tool categories:
|
|
HSM-Related Tools |
Helps you manage your HSM migration and backups more effectively |
CLEANER Tools |
Finds and resolves errors across your DASD environment that could cause you future problems |
VSAM Tools |
Helps you manage your site’s VSAM files, reducing the likelihood of space-related issues and improving I-O performance |
PDS Tools |
Provides several handy utilities, adding functionality for working with your partitioned datasets |
General-purpose Batch Utilities |
Includes useful batch utilities that no mainframe site should be without |
ONLINE Tools |
Functions available in TSO/ISPF |
This category is a collection of DMS batch utility programs to help you manage your HSM environment. It includes:
|
|
|
AHAB |
Auditor for HSM Auto-Backup |
Validates that the Auto-Backup cycle completed all critical backups |
HASP |
HSM Assistant Process |
Finds what is not getting migrated and permits external control to resolve migration issues |
HSMMON |
HSM Task Monitor |
Validates that HSM background tasks are running consistently and completing all work |
When your site has a critical auditing requirement that all updated application files must be backed-up by HSM's daily Auto-Backup cycle, HSM alone cannot meet this need. HSM can identify which datasets needed to be backed-up and perform the needed backups, but it does not guarantee that every file is backed-up appropriately. An individual dataset backup can be bypassed for several reasons, including:
- The dataset is on a non-SMS volume
- The dataset’s management class does not support Auto-Backup
- The dataset was found to be in-use during the Auto-Backup cycle and concurrent copy is not defined in the management class
- The Auto-Backup window is too narrow to accomplish all backups within the allotted time
AHAB provides the following capabilities:
- It generates reports that show which datasets are excluded from Auto-backup due to the reasons above. Armed with this information you can choose to move datasets or alter their management classes to a proper value.
- Once the issues on these reports have been resolved, daily use of AHAB provides an alert when the Auto-Backup cycle misses one or more datasets.
- If datasets were missed during auto-backup, AHAB can issue an explicit HBACKDS command to force an HSM backup on each of these datasets. This provides the certainty that the HSM backups get completed.
If you are looking for a way to reduce DASD usage at your site, HASP may be the answer.
HASP is a component of DCMS/DMS that offers a valuable tool for the z/OS storage administrator, providing reporting that shows disk files that might not be properly managed by IBM’s HSM (Hierarchical Storage Manager) and a facility that permits dynamic handling of these files. HASP looks for datasets that are not getting migrated by HSM and assigns each of these files to one of several categories.
- UNKNOWN - The datasets that are not assigned to an HSM management class
- DELAYED - The datasets that datasets should be migrating but are not
- REQUIRED - The datasets that HSM will only migrate if explicitly instructed to do so
- NON SMS - The datasets that reside on non-SMS-managed volumes and cannot be migrated
- UNSUPPORTED - The datasets that have attributes that cause them to be excluded from migration
- EXCLUDED - The datasets assigned to a management class that does not support migration.
Here are some examples of how you could use HASP:
- Find all your site’s disk datasets not accessed in more than 2 years and then force those datasets to be migrated.
- If some datasets are assigned a management class that does not support migration, change the management class to one that does.
- Find all your site’s applications files that reside on disk volumes not managed by SMS and therefore can never migrate. Selectively move some of these files to SMS-managed volumes.
- Find all your site’s empty test VSAM files not accessed for more than 5 years and selectively delete them.
- Show all the miscellaneous files assigned to a management class with auto-migration, but will never migrated due to various file attributes.
The Excel pivot chart shown below is an example of the information that HASP captures:
|
|
|
|
151,603 |
5,105,789,045 |
|
56,326 |
521,767,313 |
|
11,806 |
247,130,082 |
|
357 |
90,668,427 |
|
226 |
6,517,830 |
|
1,045 |
4,250,988 |
|
74 |
419,156 |
|
221,437 |
5,976,542,841 |
If your goal is to reduce DASD storage utilization as much as possible, HSM’s background tasks must run consistently and finish its required workload. HSMMON is a process that tracks the completion of the HSM background tasks that are associated with migration and backups. It produces a daily report that shows graphically when each HSM task has completed during the past 60 days. HSMMON provides an email alert when any of the tasks has not completed successfully for the specified number of days.
Sample HSMMON task completion graph:
|
|
|
|
|
223300000000011111111112222222222330000000001111111111222222 |
|
|
890112345678901234567890123456789011234567890123456789012345 |
|
|
WTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFSSMTWTFS |
|
SYSA |
CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC ------CCC |
|
SYSA |
CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC |
|
SYSA |
CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC --CCCCCCCCCC |
|
SYSA |
CCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCCC CCCCCCCCCCCC |
Sample HSMMON alert message: AUTOMATIC BACKUPS ON SYSA HAS NOT COMPLETED IN 3 DAYS!
CLEANER Tools are a collection of batch utility programs that help you find and eliminate errors within your DASD environment and includes:
CATCLEAN |
Match Catalog to DASD |
Finds orphaned and invalid catalog entries |
VOLCLEAN |
Match DASD to Catalog |
Finds orphaned VTOC entries |
CATSYNCH |
Synchronize Master Catalog Aliases |
Finds mismatches that cause file access errors |
After many years, a z/OS system catalog structure gradually accumulates minor errors that are difficult to detect. The problems are evident when you attempt to view a dataset in ISPF 3.4 but it shows “DATASET NOT FOUND” or “NOT MIGRATED”. The errors that are detected are:
- Dataset is defined to the catalog but does not exist on corresponding DASD volume.
- Dataset is defined to the catalog as HSM-migrated but not actually known to HSM.
Today’s modern z/OS environments use System Managed Storage (SMS) which enforces the rule that all DASD files are managed by the system catalog structure. Occasionally errors occur that cause DASD to become out of synchronization with the catalog --- a file exists but it is not cataloged. When an orphaned dataset exists on a SMS-managed volume, it is inaccessible even when referenced by its VOL=SER. These orphaned files take up space on the DASD volume. Deleting it frees the unusable space on the DASD volume.
This program can be used to compare your DASD volumes against the system catalog structure, showing uncataloged and miscataloged files.
When your mainframe environment is separated into multiple logical partitions (LPARs) within a single SYSPLEX and each of these LPARs uses a unique master catalog, it can be difficult to ensure that aliases are defined identically into each master catalog. Failure to properly synchronize the aliases can cause numerous disk issues, including files not found or duplicated data.
This program can be used validate that your master catalogs contain identical alias entries. It extracts alias information from each master catalog and reports when it finds:
- An alias defined in one master catalog but not the remaining master catalogs
- An alias pointing to a different user catalog than the alias entries in the remaining master catalogs
The VSAM batch utilities help you manage your site’s VSAM files and include:
VSAMORG |
Reorganize VSAM File |
Consolidates extents and CA/CI splits, providing faster I-O. |
VSAMTUNE |
Tune VSAM File |
Adjusts VSAM definitions to prevent space-related abends and improve I-O performance. |
This program is used to automatically reorganize the specified VSAM file. The standard IBM IDCAMS utility is used several times to accomplish the following:
- A LISTCAT function produces a report on disk that describes the VSAM dataset.
- The program reads the LISTCAT report and generates multiple sets of control statements.
- A REPRO function backs-up the VSAM dataset to either disk or tape.
- A DELETE function deletes the VSAM dataset.
- A DEFINE function re-allocates the VSAM dataset, by default exactly as it was defined previously.
- A REPRO function restores the VSAM dataset from the backup.
Here are some of the features of this facility:
- The reorganization of the specified VSAM file is entirely automatic. Just specify the DSN.
- Externally stored DEFINE statements are not used. A dynamically-generated LISTCAT report provides all the necessary information.
- The VSAM file is backed-up to disk if sufficient space is available, otherwise to tape.
- Many of the DEFINE parameters can be overridden via keywords in the execution PARM, so you can provide your own manual adjustments..
- Should the DEFINE or the restore fail, automatic recovery is attempted.
- Defines and builds each alternate indexed that is attached to the cluster
Here are the limitations of this facility:
- Supports up to 16 alternate indexes attached to the cluster
- Supports up to 8 volumes assigned to each component.
- Supports VSAM maximum record lengths up to 32752.
- Does not support KEYRANGES.
This program is used to automatically tune and reorganize one or more VSAM files. A large part of application performance can be attributed to the speed at which data can be stored and retrieved from its VSAM files. A poorly tuned VSAM file can greatly increase its storage/retrieval time. Unfortunately, IBM’s manuals do little to educate the casual reader on the subject of VSAM tuning. This program was developed to greatly simplify the tuning process.
Internally using the DMS VSAMORG utility to perform the actual reorganization of the VSAM file, this program can “tune” the VSAM file by adjusting selected parameters as the VSAM file is being re-DEFINED.
While processing each VSAM file, the IBM IDCAMS utility is used several times to accomplish the following:
- A LISTCAT function produces a report on disk that describes the VSAM dataset.
- The program reads the LISTCAT report and compares LISTCAT values to the corresponding thresholds.
- If threshold criteria are not met, processing terminates without any further processing.
- The dataset name and tuning values are passed to VSAMORG to reorganize the dataset:
- The program reads the LISTCAT report and generates multiple sets of control statements.
- A REPRO function backs-up the VSAM dataset to either disk or tape.
- A DELETE function deletes the VSAM dataset.
- A DEFINE function re-allocates the VSAM dataset, using modified “tuned” parameters.
- A REPRO function restores the VSAM dataset from the backup.
Here are several features of this facility:
- The reorganization and tuning of each specified VSAM file is entirely automatic.
- Externally stored DEFINE statements are not used. A dynamically generated LISTCAT report provides all the necessary information.
- The VSAM file is backed-up to disk if sufficient space is available, otherwise to tape.
- TThe thresholds that determine if a file is reorganized are defined externally.
- Should the DEFINE or the restore fail, automatic recovery can be attempted.
The PDS batch utilities add useful functions to help manage your partitioned datasets and include:
PDSMON |
PDS Space Monitor Report |
Alerts you when a listed PDS has exceeded full threshold |
PDSSTOR |
Store PDS Member w/Directory Entry |
Permits you to create PDS members that are stored with a TSO-type directory entry |
PDSDEL |
Delete PDS Member |
Safe delete of library members using DISP=SHR |
PDSUPD |
Update PDS Directory |
A batch equivalent to ISPF option 3.5 |
This program produces a report that shows the percentage of data and directory blocks in use on each of the specified PDS or PDS-E libraries. One print line is generated for each library. When one or more libraries exceeds the space-used thresholds (data or directory), an error message is presented on the report line and the program terminates with a return code 12. The supplied sample JCL cause an email alert to be sent when this happens.
The libraries you should define to the process are the ones that change frequently. Typically, the SYSIN file defines each of your site’s application test, staging and production library. These libraries occasionally become full, causing issues and delays as the library must be expanded before work can continue. Using this utility, you can be aware of libraries that are approaching full and can be more proactive about resolving space issues before they happen.
This program permits the user to store the input file as a member on a partitioned dataset (PDS) and cause a TSO-type entry to be updated in the directory of the PDS.
System enqueues are issued on the updated library to prevent corruption that could be caused by simultaneous tasks that are updating the same library using this utility, the system Binder (linkage-editor) or the ISPF Editor. The library can safely be updated with DISP=SHR.
This program permits deletion of selected PDS members while the library is allocated DISP=SHR. You can process a single member specified in the execution PARM or a list of members identified in the SYSIN file.
System enqueues are issued on the updated library to prevent corruption that could be caused by simultaneous tasks that are updating the same library using this utility, the system Binder (linkage-editor) or the ISPF Editor. The library can safely be updated with DISP=SHR.
This program updates the directory entries of one or more PDS members. Using the execution PARM, you can:
- Choose the desired member names to be updated, optionally using wildcards
- Specified the desired directory version and modification levels
- Specified the desired directory last-changed user ID
- Set the current date at the last-changed date
System enqueues are issued on the updated library to prevent corruption that could be caused by simultaneous tasks that are updating the same library using this utility, the system Binder (linkage-editor) or the ISPF Editor. The library can safely be updated with DISP=SHR.
These batch utilities don’t fall into one of the other categories, but add useful functionality. They include:
GSIDCAMS |
IDCAMS Front-End |
Execute GSIDCAMS instead of IDCAMS to improve REPRO I-O performance |
DYNVOL |
Dynamic Volume Utility |
Simplify your jobsteps that need current volume names |
TESTDS |
Test Existence of Dataset |
Use to conditional execute a jobstep when a dataset exists |
CONSCMD |
Console Commander |
Process an input file of MVS and JES commands |
Use GSIDCAMS in place of standard IBM utility IDCAMS to cause a jobstep abend whenever IDCAMS terminates with a ‘higher-than-expected’ return code. To accomplish this, GSIDCAMS simply calls IDCAMS and, upon IDCAMS’s completion, tests its return code. Based on the execution PARM, it decides whether an abend is warranted.
Another nice function of GSIDCAMS is to internally convert IDCAMS REPRO operations to use a SORT COPY function. SORT uses EXCP processing for its COPY operation, much faster and more efficient than the QSAM processing used by IDCAMS during a REPRO. GSIDCAMS improves performance on any IDCAMS REPRO step that uses a sequential file for input and/or output operations.
In a benchmark test, both utilities were used to REPRO a 5-million record sequential file. The following sample output shows the marked difference in performance between using IDCAMS (the second step) and GSIDCAMS (the third step). GSIDCAMS used less than 1% of the EXCP count (physical I-O’s) compared to IDCAMS, causing a reduction in CPU time (TCB column) and wall-clock time.
19.30.24 JOB02601 IEF403I HXGDO0ID - STARTED - TIME=19.30.24 |
19.30.35 |
JOB02601 |
-STEPNAME |
PROCSTEP |
RC |
EXCP |
CONN |
TCB |
SRB |
CLOCK |
19.30.35 |
JOB02601 |
-TESTDATA |
|
00 |
17389 |
4381 |
.05 |
.00 |
.1 |
19.30.46 |
JOB02601 |
-IDCAMS |
|
00 |
34535 |
6878 |
.02 |
.00 |
.1 |
19.30.50 |
JOB02601 |
-GSIDCAMS |
|
00 |
327 |
5678 |
.00 |
.00 |
..0 |
19.30.50 JOB02601 IEF404I HXGDO0ID - ENDED - TIME=19.30.50 |
This handy utility eliminates the need to maintain lists of disk VOLSER's for jobs that must produce VTOC listings and for other types of DASD processing.
Without GSDM0105, each time DASD is added to the system, one or more DASD management jobs must be changed to include the new VOLSER's. Likewise, when DASD is removed, these jobs must be changed again.
GSDM0105 reads the UCB chain in the operating system to automatically include all or generically selected VOLSER's. It can dynamically allocate UNIT statements required by many IBM utilities. It can also generate utility control statements from a model file, substituting VOLSER's in their proper position.
This program accepts a dataset name in the execution PARM and passes a return code based on whether the dataset is cataloged. This utility can be used when a batch step needs to be conditionally executed based on the existence of a dataset.
This handy utility permits an input file of MVS operator commands to be processed by the operating system. The commands are issued when the step is executed rather than submit-time when specified as instream JCL.
The ISPF-based DMS utilities work without requiring any changes to the ISPF environment. They provide quick access to several handy tools, including:
SPACE |
DASD Volume Display |
Shows mounted disk volumes in a sortable, actionable list |
WHOHAS |
List Users of a Dataset |
Shows the tasks that are enqueued on a dataset |
LCB |
LISTCAT Browse Utility |
Use as line command in ISPF 3.4 to browse LISTCAT report on associated dataset |
LCS |
LISTCAT Screen Utility |
Use as line command in ISPF 3.4 to display LISTCAT on associated dataset |
MENU |
DMS Online Function Menu |
Provides access to the functions above from an ISPF menu |
DMS is available for any processor that supports the z/OS operating system environment. DMS utilizes ISPF/Dialog Manager (DM) and ISPF/Program Development Facility (PDF). DMS supports all versions and releases of z/OS.
The DCMS Customer Support Hotline (952)473-3572 is accessible 24 hours / 7 days a week. On-site training and consulting are available for a nominal fee.
The DMS system can be downloaded from the
DCMS website for a no cost, no obligation trial. The package includes upload-able binary files and documentation PDF’s.
Products are licensed (Perpetual and Renewable) on a per processor basis. Tiered pricing is based on Processor Software Groups. Annual maintenance for perpetual licenses is calculated at a rate of 15% of the current license fee.
- DCMS DMS Administrator Guide
- DCMS DMS Upgrade Guide