VSAM Interview Questions

VSAM (Virtual Storage Access Method) is both a dataset type and an access method for managing various types of datasets. When used as an access method, VSAM offers greater functionality with greater performance and flexibility than other disk access methods. In VSAM, records are maintained in a unique format that is not understandable by other methods. VSAM organizes data as files on Mainframe computers. This is one of the most common high-performance file access methods used in different versions of operating systems such as MVS, z/OS, and OS/390. With VSAM, enterprises can organize records in a file in physical order (the order they were entered) or logical order using a key (for example, the employee ID number) or based on their relative record numbers on DASD (Direct Access Storage Device). Record lengths can be fixed or variable in VSAM.

In order to convert flat files into VSAM, follow the steps below:

• Step1: Decide what is going to be your primary key. Alternatively, you can use the entire record as your primary key.
• Step2: Sort the file by the primary key, or the entire record if no primary key is specified. Drop duplicates using a SORT statement.
• Step3: Next, you must run a JCL command (IDCAMS - Define Cluster) with step-2 results/output (sorting mechanisms) as input.
• Step4: Finally, to load the flat file into VSAM, use the IDCAMS REPRO command.

A VSAM file that contains no records or an empty VSAM file cannot be opened for the purpose of being used as an input to the COBOL program. Users will no longer be able to access them. Rather, an empty file may be opened for output purposes. COBOL writes a dummy record into the file and deletes it once the user opens it.

VSAM Share Options determine how users can access VSAM datasets. By using these options, different users/jobs can be granted different levels of access to a single VSAM dataset. As part of the DEFINE statement, the SHAREOPTS parameter is written as SHAREOPTS (a,b). Here, a represents the cross-region share option, which indicates how two or more users/jobs on the same MVS system can share a file, and b represents the cross-system share option, which indicates how two or more users/jobs on different MVS systems can share a file.  In most cases, this value is  SHAREOPTS(2 3).

If the cross-region value is 2, multiple users can simultaneously process the file provided that only one of them is an updater. The value 3 for the cross-system indicates the file can be processed by any number of jobs/users (VSAM does not ensure integrity).

There are a number of scenarios in which a VSAM file would be preferred over a Sequential file. Here are a few:

• When performing multiple operations on a file based on a specific key value.
• When performing multiple operations on a file, such as updating a specific record or removing files.
• Whenever you need a file whose contents will need to be modified (update, delete, insert records) frequently.
• Where multiple operations are required like different ways of accessing records either sequentially, with a key, or dynamically.

VSAM (Virtual Storage Access Method) and Non-VSAM files differ in the following ways:

You can refresh the VSAM file once a day, every week, or every month, however, you prefer. Refreshing is the same as deleting/defining the VSAM file.

• Step 1: Delete existing/current VSAM file

 IF LASTCC =8 ; SET MAXCC=0

• Step 2: Create a new VSAM File

 DEFINE VSAM-FILE_NAME (COND=(0,LT,STEP1)

Whenever you try to delete a file that doesn't exist, you get a return code of 8. The command VERIFY can be used in TSO Prompt to verify that the file exists.

"Freespace" is an important parameter in "Define cluster." In this parameter, we specify the percentage of freespace (available storage) that will be reserved for the CI (control intervals) and CA (control areas) of the data component. The default value is 0 per cent.

Syntax:

FREESPACE (CI-percent, CA-percent)

ESDS files store records in the order they were added to the dataset. The records are referenced by their physical address, also known as RBA (Relative Byte Address). Assuming we have 60 bytes per record in an ESDS dataset, then RBA of the 1st record would be 0, RBA of the 2nd record would be 60, RBA of the 3rd record would be 120, and so on. RBA allows for sequential access to records, known as addressed access. The data is stored in the order in which it was entered. New records are added at the end. Records cannot be deleted from the ESDS dataset, but they may be marked as inactive. ESDS dataset records can be of variable or fixed length.

Both QSAM and ESDS files store records in sequential order, i.e., according to the entry order. However, they differ in several ways:

Listed below are the common file status codes:

• 00: The operation was successful.
• 04: Fixed length record was invalid.
• 20: The key for KSDS or RRDS is invalid.
• 22: Duplicate primary key found.
• 23: Error locating the record or file.
• 30: Permanent I/O fault.
• 38: Attempting to open a locked file.
• 91: Invalid password or authorization.
• 93: Resources are unavailable.
• 95: Invalid or incomplete file information.
• 98: File Locked- Failed to open the file.
• 99: Record Locked - Failed to access record, etc.

As you work with VSAM datasets, you may encounter abends during the process. You can find the status of a VSAM file by its VSAM file status code. VSAM files need to have at least one data record loaded into them before they can be opened for input or updating.

VSAM data sets can be classified into the following types:

• RRDS (Relative-record data set): It contains records ordered by relative record number and is accessible via this number. In RRDS datasets, records are stored in slots of a fixed length.
• ESDS (Entry-sequenced data set): It contains records in the order of entry or in sequential order and is accessible only via relative byte addresses. ESDSs are similar to sequential data sets. Records in ESDS datasets can have variable or fixed lengths.
• KSDS (Key-sequenced data set): It contains records ordered by key field and is accessible via key-value or relative byte address. Records in KSDS dataset can be of fixed length or variable length.
• LDS (Linear data set): It is composed of an unstructured data set whose Control Interval (CI) size is multiple of 4K.

In contrast to non-VSAM data sets, VSAM data sets are formatted differently. The VSAM data sets consist of collections of records that are grouped into CI (control intervals). VSAM stores its records within the control interval, which is a fixed portion of storage space. The maximum length of a VSAM record is 1 cylinder. In VSAM, records are arranged according to the index key, the relative record number or the relative byte address.

Each file in VSAM has two objects, each of which defines its storage system.

• File object: It represents a single file. File methods such as Read, Delete, Rewrite, Write, Unlock, and StartBr can be utilized to perform these actions.
• FileBrowse object: This object represents browse operations done on a file. It involves Enable, ReadNext, ResetBr, and ReadPrv methods, and these are performed using IDCAMS or through TSO prompts.

IDCAMS is an acronym for Integrated Data Cluster Access Method Services. VSAM datasets can be easily manipulated with IDCAMS Utility. It can be utilized for creating, deleting, and modifying VSAM datasets.

An AMS (Access Method Service) is a service program that performs a wide range of functions on VSAM/Non-VSAM datasets and catalogues. AMS includes a powerful utility program known as IDCAMS that can be invoked via a standard JCL and is able to carry out a slew of AMS functions through its commands. AMS deploys two kinds of commands:

• Functional commands: These are commands that demand real action, like DEFINE CLUSTER, REPRO, VERIFY, DELETE, BLDINDEX, and LISTCAT.
• Modal commands: These are commands that specify conditions for executing functional commands such as "IF-THEN-ELSE". A TSO (Time Sharing Option) user can only use functional commands, not modal commands.

Generation(0) refers to the current GDG generation, Generation(-1) refers to the GDG generation just prior to the current generation, and Generation(+1) refers to the GDG generation following the current generation.

• When a GDG exists in 3 generations, then the 3rd generation is the current generation, and therefore, we refer to it using generation(0) in JCL.
• The generation (2nd generation) right before the current generation (3rd generation) is referenced with generation(-1) in JCL.
• Lastly, the generation (1st generation) just before the 2nd generation is referenced using generation(-2) in JCL.

Conclusion

In the 1970s, VSAM was introduced to replace the older IBM file-access methods with greater functionality, and ease-of-use, and to reduce device-dependency issues. The organized index in VSAM makes it easy to access data, and it also supports many datasets within one cluster. For those preparing for an interview in this niche, the above-listed VSAM interview questions will be beneficial to both experienced and new candidates. 
With this collection of 30+ solved VSAM questions, you can prepare for technical interviews as well as online selection tests during campus placements and interviews. Having this arsenal of tricky VSAM questions in hand, you'll be able to handle objective type and multiple-choice questions on this topic with ease.

Learning VSAM will be easier if you are conversant with COBOL, although it is not required. The attentive interviewer rarely plans to ask you a specific question during the interview; instead, they start by asking some basic questions about the subject, followed by further questions based on the discussion and your response. Candidates should have solid conceptual knowledge, as well as the ability to communicate confidently and clearly. It's a great way to build confidence and prepare for various interview scenarios by taking a mock interview. Good preparation and a bit of cleverness will easily get you through this interview.

Useful Resources

https://www.interviewbit.com/jcl-interview-questions/
https://www.interviewbit.com/cobol-interview-questions/
https://www.interviewbit.com/mainframe-interview-questions/
https://www.interviewbit.com/technical-interview-questions/

For identifying a specific generation of a GDG, an absolute generation and version number are used. GnnnnVmm is the format of the generation and version numbers.

Syntax:

ABC.PQR.XYZ.GnnnnVmm

Here,

• XYZ.ABC.PRS.GDG represents GDG Base.
• Gnnnn represents the Generation number (Value range: 0000 – 9999).
• Vmm represents Version number (Value range: 00 – 99).

Example:

If the GDG base name is SCALER.DATA.SAMPLE.GDG, then the datasets should be named as follows:

• SCALER.DATA.SAMPLE.GDG.G0001V00
• SCALER.DATA.SAMPLE.GDG.G0002V01, and so on.

Here, 

• SCALER.DATA.SAMPLE.GDG.G0001V00 is the generation data set 1, version 0 in the generation data group SCALER.DATAI.SAMPLE.GDG.
• SCALER.DATA.SAMPLE.GDG.G0002V01 is the generation data set 2, version 1, in the generation data group SCALER.DATA.SAMPLE.GDG.

The latest/current version of the GDG is SCALER.DATI.SAMPLE.GDG(0). The current generation of GDG is called generation(0), previous versions are called SCALER.DATA.SAMPLE.GDG(-1), and so on. SCALER.DATA.SAMPLE.GDG(+1) is the next generation to be created after the current generation.

Generally, a GDG (Generation Data Group) comprises a collection of related non-VSAM data sets that are chronologically arranged. These datasets are grouped together based on a common name known as the GDG base, and each dataset associated with the GDG base is called a GDG version. A GDG base can have up to 255 generations. They are highly effective tools for processing data sets created on a periodic basis, i.e., monthly, weekly, daily, and hourly. Each processing cycle results in a new generation of the dataset. GDG can be used for a myriad of purposes, such as Invoicing, Statistics, Reporting, Back-Up, Audit Trails, etc.

To create KSDS dataset, IDCAMS DEFINE CLUSTER command is used with INDEXED parameters. A few other important parameters are RECORDSIZE, KEYS, etc.

Syntax: Here is the JCL to create a KSDS dataset with minimum parameters.

DEFINE CLUSTER (NAME(ksds-file-name) -
VOLUMES(volume-serial) -
INDEXED -
RECORDSIZE (average maximum) -
[FREESPACE(CI-Percentage,CA-Percentage)] -
KEYS(length, offset) -
CISZ(number) -
DATA - (NAME(ksds-file-name.data)) -
INDEX - (NAME(ksds-file-name.index))

Here,

• INDEXED Parameter indicates that the created file is KSDS.
• All remaining parameters are almost the same as in ESDS or KSDS.

The following three modes of access are available for VSAM files:

• Sequential Access: Enter the code ACCESS IS SEQUENTIAL in the FILE-CONTROL entry for sequential access. Records within indexed files can be accessed based on the key field selected (either the primary key or an alternate key), and records within relative files can be accessed based on their relative record numbers.
• Random Access: Enter the code ACCESS IS RANDOM in the FILE-CONTROL entry for random access. Records within indexed files can be accessed based on the value entered in a key field, and records within relative files can be accessed based on the value entered in the relative key.
• Dynamic Access: Enter the code ACCESS IS DYNAMIC in the FILE-CONTROL entry for dynamic access. In dynamic access, sequential and random access are both combined within the same program. By using dynamic access, a single program can be written to perform both sequential and random operations, allowing some records to be accessed in sequential order and others based on their keys.

Using the REPRO command, data can be loaded into a VSAM dataset. This command can also be used to copy data between VSAM data sets, as well as to copy data from sequential files to VSAM files. The IDCAMS utility loads datasets using the REPRO command.

Syntax:

REPRO INFILE(in-ddname)
OUTFILE(out-ddname)

Here, 

• in-ddname represents the DD name of the input dataset which has records.
• out-ddname represents the DD name of the Output Dataset, into which the input dataset records will be copied.

The steps for creating an Alternate Index are as follows:

Step1: Define Alternate Index

DEFINE ALTERNATEINDEX command is used to define an alternate index.

Syntax:

DEFINE ALTERNATEINDEX -
(NAME(alternateindex-name) -
RECORDSIZE(average maximum)) -
FREESPACE(CI-Percentage,CA-Percentage) -
RELATE(vsamfile-name) -
UPGRADE / NOUPGRADE -
KEYS(length offset) -
CISZ(number) -
NONUNIQUEKEY / UNIQUEKEY -
DATA - (NAME(vsamfile-name.data)) -
INDEX - (NAME(vsamfile-name.index))

Here,

• DEFINE ALTERNATEINDEX define Alternate Index.
• NAME specifies the Alternate Index name.
• RELATE specifies the VSAM cluster name for which the alternate index is created.
• NONUNIQUEKEY / UNIQUEKEY; NONUNIQUEKEY signifies that duplicate entries may exist for the alternative index, while UNIQUEKEY indicates the alternate index is unique.
• UPGRADE / NOUPGRADE; NOUPGRADE specifies that an alternate index shall be unchanged if the base cluster is changed, while UPGRADE specifies that it shall be modified if the base cluster is changed.

Step2: Define Path

Using the Define Path, the alternate index can be related to the base cluster. Defining a path includes specifying its name and the alternative index it is related to.

DEFINE PATH -
NAME(alternateindex-path-name) -
PATHENTRY(alternateindex-name))

Here,

• NAME specifies the Alternate Index Path Name
• PATHENTRY Specifies Alternate Index Name.

Step3: Building Index

The alternate index is built using the BLDINDEX command. It reads the entire VSAM indexed data set (or base cluster) and extracts essential data to build the alternate index.

BLDINDEX -
INDATASET(vsamcluster-name) -
OUTDATASET(alternateindex-name))

Here,

• INDATASET specifies the VSAM Cluster Name
• OUTDATASET specifies the Alternate Index Name.

Additional indexes are created for ESDS/KSDS datasets in addition to the primary index. Alternate indexes enable records to be accessed by using more than one key. The key of the alternate index is not necessarily unique; it can have duplicates.

A number of operations can be carried out on VSAM datasets using VSAM commands. Here are some of the most useful VSAM commands:

• Alter: Modifies VSAM file attributes (cluster, path, catalogue, etc.)
• Repro: Loads the VSAM dataset with data.
• Listcat: Obtains VSAM dataset catalogue details.
• Verify: Checks and fixes VSAM files that have not been properly closed.
• Print: Prints VSAM dataset’s content.
• Delete: Delete VSAM and non-VSAM datasets, as well as the catalogues.

If VSAM is running out of space, you can fix the issue as follows:

• Create a new VSAM dataset with additional space allocated.
• Copy the old VSAM file into the new VSAM dataset using the IDCAMS REPRO command.
• Alter/rename/delete an old VSAM dataset using IDCAMS.
• With IDCAMS, the new VSAM dataset can be altered/renamed to the original VSAM dataset's name.

The Control Interval (CI) size should depend on the length of the record and the type of processing. In most cases, CI is 4K. In cases where the record length exceeds 1K, choose CI sizes of 6K or 8K. If more insertions are anticipated, then FREESPACE (CI-percent, CA-percent) should be large. Set the value to (20-20) when updates are heavy.

CI and CA split occur when new data records are inserted. When a record is to be inserted in a CI and there is not enough space to fit it, the CI will be split.

• CI SPLIT: Essentially, it refers to the movement/transfer of some records from an existing CI to a free CI within the same CA when an addition or update is not feasible in the existing CI.
• CA SPLIT: It refers to the movement/transfer of half of the records from one CA to another one when an addition or update is not feasible in the existing CI.  This results in two CAs that are around half full rather than one full and one empty.

In the data component of the VSAM cluster, each data record is stored as a part of a unit called the Control Interval (CI), and each Control Interval is stored in a unit called the Control Area (CA).

• Control Interval (CI): It is a contiguous storage area that VSAM uses to store data records and other control information. A block represents a unit of data being transferred between memory and storage devices in non-VSAM data management methods. CI presents a unit of data transferred between the storage device (DASD) and memory (buffer) in VSAM. A control interval size is specified at the time the cluster is defined. The size of Control Intervals is measured in bytes. For example, 2 kilobytes.
• Control Area (CA): A control area consists of several CI (Control intervals). Depending on the number of records, a VSAM dataset consists of one or more CAs. Data records are grouped into CIs, and Control Intervals are grouped into CAs. Sizes of Control Areas are measured in cylinders or disk tracks.

The VSAM consists of datasets used to store the data and are referred to as clusters. Index, sequence set, and data portion of a dataset are all grouped into clusters. Each VSAM cluster occupies an area divided into contiguous areas called CI (Control Intervals). A VSAM cluster consists of two main components:

• Index Component: This component consists of the index part. VSAM retrieves records from the data component using the index component.
• Data Component: This component contains the data. It contains the actual data records.

The IDCAMS utility can process more than one AMS command in a single execution. Thus, AMS sets a condition code after every command execution and also a maximum condition code value for each execution. The syntax for the AMS Modal command is given below. Only IDCAMS can be used for this purpose and not the TSO prompt.

Syntax:

IF {LASTCC | MAXCC} {operator} {numeric value}
    THEN {command} |
        DO {command set}   //(such as DEFINE, PRINT, DELETE, etc...)
        END
[ELSE {command} |
        DO
          {command set}
        END]

Here, in the IF statement, the condition code to be tested is specified as either LASTCC or MAXCC.

• The LASTCC is the condition code from the last command that was executed (such as delete) just before the IF structure was executed.
• A MAXCC is a condition code returned by any of the previous functions.
• Operators can be GT or >, LT or <, EQ or = , NE or ¬= (not equal to), GE or >= , LE ro <=.