What is KSDS in VSAM?

KSDS (Key Sequenced Data Set) is a VSAM file organization that stores records in key sequence and provides indexed access by a primary key. Records are automatically maintained in sorted order by key, and you can access records directly by key value or sequentially. KSDS supports both unique and duplicate keys and provides efficient random and sequential access. KSDS is the most commonly used VSAM organization type for applications requiring key-based access.

What is ESDS in VSAM?

ESDS (Entry Sequenced Data Set) is a VSAM file organization that stores records in the order they are written, similar to sequential files. Records are accessed sequentially in the order they were written, and you cannot delete or update records in place (only add new records). ESDS is useful for log files, audit trails, and applications where record order and append-only operations are important.

What is RRDS in VSAM?

RRDS (Relative Record Data Set) is a VSAM file organization that provides direct access by relative record number (RRN). Each record has a fixed-length slot identified by its position (1, 2, 3, etc.). You can read, write, update, or delete records by specifying the relative record number. RRDS is useful when you have a known, fixed set of records that can be accessed by position, such as lookup tables or fixed-size arrays stored on disk.

How do you define a VSAM file in COBOL?

You define a VSAM file in COBOL using SELECT and FD statements. In FILE-CONTROL: SELECT file-name ASSIGN TO dataset-name ORGANIZATION IS INDEXED (for KSDS), ORGANIZATION IS SEQUENTIAL (for ESDS), or ORGANIZATION IS RELATIVE (for RRDS). For KSDS, specify ACCESS MODE IS RANDOM or SEQUENTIAL, RECORD KEY IS key-field, and optionally ALTERNATE RECORD KEY. In FILE SECTION, define the FD and record structure. VSAM files must be created and cataloged before use.

What is the difference between KSDS, ESDS, and RRDS?

KSDS provides indexed access by key with automatic sorting, supports random and sequential access, and allows updates and deletes. ESDS stores records in write order, supports only sequential access, and is append-only (no updates or deletes). RRDS provides direct access by relative record number, supports random access by position, and allows updates and deletes. Choose KSDS for key-based access, ESDS for sequential logs, and RRDS for position-based access.

How do you access VSAM files in COBOL?

You access VSAM files using standard COBOL I/O statements: OPEN (INPUT, OUTPUT, I-O, or EXTEND), READ, WRITE, REWRITE, DELETE, and CLOSE. For KSDS with RANDOM access, set the RECORD KEY before READ or WRITE. For SEQUENTIAL access, use READ NEXT. For RRDS, set the relative record number before READ or WRITE. VSAM handles the underlying access method automatically based on the file organization and access mode specified.

What is a RECORD KEY in VSAM KSDS?

RECORD KEY in VSAM KSDS is the field used to identify and access records. It must be defined in the record structure and specified in the SELECT statement. The RECORD KEY must be unique (or allow duplicates if specified). For random access, you set the RECORD KEY to the desired value, then READ or WRITE. VSAM uses the key to locate records in the index. The key field determines the sort order of records in the file.

MainframeMaster

COBOL Tutorial

Progress0 of 0 lessons

COBOL VSAM Concepts

Q: What is VSAM in COBOL?

VSAM (Virtual Storage Access Method) is an IBM mainframe file access method that provides efficient, indexed access to data. VSAM offers three main organization types: KSDS (Key Sequenced Data Set) for indexed access by key, ESDS (Entry Sequenced Data Set) for sequential access, and RRDS (Relative Record Data Set) for access by relative record number. VSAM provides better performance, indexing, and data management capabilities compared to traditional sequential files.

VSAM (Virtual Storage Access Method) is IBM's mainframe file access method that provides efficient, indexed access to data. VSAM offers three main organization types: KSDS (Key Sequenced Data Set) for indexed access by key, ESDS (Entry Sequenced Data Set) for sequential access, and RRDS (Relative Record Data Set) for access by relative record number. Understanding VSAM concepts is essential for working with mainframe files that require indexed access, efficient retrieval, and advanced data management capabilities in COBOL programs.

What is VSAM?

VSAM is an IBM mainframe file access method that provides:

Indexed access: Direct access to records by key value
Automatic sorting: Records maintained in key sequence
Efficient retrieval: Fast access using indexes
Multiple organization types: KSDS, ESDS, and RRDS
Data integrity: Built-in recovery and consistency features
Space management: Automatic space allocation and management

VSAM files must be created and cataloged before use in COBOL programs.

VSAM Organization Types

VSAM provides three main organization types:

VSAM Organization Types
Type	Full Name	Access Method	Best Use
KSDS	Key Sequenced Data Set	Indexed by key, random or sequential	Key-based access, updates, deletes
ESDS	Entry Sequenced Data Set	Sequential only, in write order	Logs, audit trails, append-only
RRDS	Relative Record Data Set	By relative record number (position)	Fixed-size records, position-based access

KSDS (Key Sequenced Data Set)

KSDS is the most commonly used VSAM organization, providing indexed access by a primary key:

KSDS Characteristics

Indexed access: Records accessed by key value
Automatic sorting: Records maintained in key sequence
Random or sequential: Supports both access modes
Updates and deletes: Can modify and remove records
Unique or duplicate keys: Supports both key types
Alternate keys: Can define alternate indexes

Defining KSDS in COBOL

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
       ENVIRONMENT DIVISION.
       INPUT-OUTPUT SECTION.
       FILE-CONTROL.
           SELECT CUSTOMER-FILE ASSIGN TO CUSTOMER.VSAM
               ORGANIZATION IS INDEXED
               ACCESS MODE IS RANDOM
               RECORD KEY IS CUSTOMER-ID
               FILE STATUS IS CUSTOMER-STATUS.
       
       DATA DIVISION.
       FILE SECTION.
       FD  CUSTOMER-FILE.
       01  CUSTOMER-RECORD.
           05  CUSTOMER-ID      PIC 9(5).
           05  CUSTOMER-NAME    PIC X(30).
           05  CUSTOMER-BALANCE PIC 9(8)V99.
           05  CUSTOMER-ADDRESS PIC X(50).
       
       WORKING-STORAGE SECTION.
       01  CUSTOMER-STATUS     PIC XX.
           88  FILE-OK          VALUE '00'.
           88  RECORD-NOT-FOUND VALUE '23'.
       
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN I-O CUSTOMER-FILE
           
           IF NOT FILE-OK
               DISPLAY "Error opening file: " CUSTOMER-STATUS
               STOP RUN
           END-IF
           
           *> Random access by key
           MOVE 12345 TO CUSTOMER-ID
           READ CUSTOMER-FILE
               INVALID KEY
                   DISPLAY "Customer not found: " CUSTOMER-ID
               NOT INVALID KEY
                   DISPLAY "Customer: " CUSTOMER-NAME
                   DISPLAY "Balance: " CUSTOMER-BALANCE
           END-READ
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

Key points for KSDS:

ORGANIZATION IS INDEXED: Specifies KSDS organization
ACCESS MODE: RANDOM for direct access, SEQUENTIAL for sequential access
RECORD KEY: The field used for indexing (must be in record structure)
Set key before READ: For RANDOM access, set RECORD KEY before reading

KSDS Sequential Access

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
       SELECT CUSTOMER-FILE ASSIGN TO CUSTOMER.VSAM
           ORGANIZATION IS INDEXED
           ACCESS MODE IS SEQUENTIAL
           RECORD KEY IS CUSTOMER-ID
           FILE STATUS IS CUSTOMER-STATUS.
       
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN INPUT CUSTOMER-FILE
           
           PERFORM UNTIL END-OF-FILE
               READ CUSTOMER-FILE
                   AT END
                       SET END-OF-FILE TO TRUE
                   NOT AT END
                       DISPLAY "Customer: " CUSTOMER-NAME
                       *> Process record
               END-READ
           END-PERFORM
           
           CLOSE CUSTOMER-FILE
           STOP RUN.

Sequential access reads records in key order, from lowest to highest key value.

ESDS (Entry Sequenced Data Set)

ESDS stores records in the order they are written, similar to sequential files:

ESDS Characteristics

Write order: Records stored in order written
Sequential access only: Cannot access records randomly
Append-only: Cannot update or delete existing records
No keys: No indexing or key-based access
Useful for logs: Ideal for audit trails and log files

Defining ESDS in COBOL

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
       ENVIRONMENT DIVISION.
       INPUT-OUTPUT SECTION.
       FILE-CONTROL.
           SELECT LOG-FILE ASSIGN TO LOG.VSAM
               ORGANIZATION IS SEQUENTIAL
               ACCESS MODE IS SEQUENTIAL
               FILE STATUS IS LOG-STATUS.
       
       DATA DIVISION.
       FILE SECTION.
       FD  LOG-FILE.
       01  LOG-RECORD.
           05  LOG-DATE        PIC 9(8).
           05  LOG-TIME        PIC 9(6).
           05  LOG-MESSAGE     PIC X(100).
       
       WORKING-STORAGE SECTION.
       01  LOG-STATUS         PIC XX.
       
       PROCEDURE DIVISION.
       MAIN-PARA.
           *> Open for output (append)
           OPEN EXTEND LOG-FILE
           
           *> Write log entry
           ACCEPT LOG-DATE FROM DATE YYYYMMDD
           ACCEPT LOG-TIME FROM TIME
           MOVE "Application started" TO LOG-MESSAGE
           WRITE LOG-RECORD
           
           CLOSE LOG-FILE
           
           *> Read sequentially
           OPEN INPUT LOG-FILE
           PERFORM UNTIL END-OF-FILE
               READ LOG-FILE
                   AT END SET END-OF-FILE TO TRUE
                   NOT AT END
                       DISPLAY LOG-MESSAGE
               END-READ
           END-PERFORM
           
           CLOSE LOG-FILE
           STOP RUN.

ESDS is useful for log files, audit trails, and any application where records must be preserved in write order and accessed sequentially.

RRDS (Relative Record Data Set)

RRDS provides direct access by relative record number (position):

RRDS Characteristics

Position-based access: Access records by relative record number (1, 2, 3, etc.)
Fixed-length slots: Each record has a fixed position
Random access: Direct access by record number
Updates and deletes: Can modify and remove records
Fixed record length: All records must be the same length

Defining RRDS in COBOL

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
       ENVIRONMENT DIVISION.
       INPUT-OUTPUT SECTION.
       FILE-CONTROL.
           SELECT LOOKUP-TABLE ASSIGN TO LOOKUP.VSAM
               ORGANIZATION IS RELATIVE
               ACCESS MODE IS RANDOM
               RELATIVE KEY IS RECORD-NUMBER
               FILE STATUS IS LOOKUP-STATUS.
       
       DATA DIVISION.
       FILE SECTION.
       FD  LOOKUP-TABLE.
       01  LOOKUP-RECORD.
           05  STATE-CODE      PIC X(2).
           05  STATE-NAME      PIC X(20).
           05  STATE-TAX-RATE  PIC V999.
       
       WORKING-STORAGE SECTION.
       01  RECORD-NUMBER      PIC 9(4).
       01  LOOKUP-STATUS      PIC XX.
       
       PROCEDURE DIVISION.
       MAIN-PARA.
           OPEN I-O LOOKUP-TABLE
           
           *> Access record at position 5
           MOVE 5 TO RECORD-NUMBER
           READ LOOKUP-TABLE
               INVALID KEY
                   DISPLAY "Record not found at position: " RECORD-NUMBER
               NOT INVALID KEY
                   DISPLAY "State: " STATE-NAME
                   DISPLAY "Tax Rate: " STATE-TAX-RATE
           END-READ
           
           *> Write to position 10
           MOVE 10 TO RECORD-NUMBER
           MOVE 'TX' TO STATE-CODE
           MOVE 'TEXAS' TO STATE-NAME
           MOVE 0.0825 TO STATE-TAX-RATE
           WRITE LOOKUP-RECORD
               INVALID KEY
                   DISPLAY "Error writing record"
           END-WRITE
           
           CLOSE LOOKUP-TABLE
           STOP RUN.

RRDS uses RELATIVE KEY (a numeric field) to specify which record position to access. Record numbers start at 1.

VSAM File Operations

VSAM files use standard COBOL I/O statements with some VSAM-specific considerations:

Opening VSAM Files

cobol

1
2
3
4
5
6
7
8
9
10
11
       *> INPUT - Read only
       OPEN INPUT VSAM-FILE
       
       *> OUTPUT - Write only (creates new file)
       OPEN OUTPUT VSAM-FILE
       
       *> I-O - Read and write (update existing file)
       OPEN I-O VSAM-FILE
       
       *> EXTEND - Append to ESDS
       OPEN EXTEND VSAM-FILE

Reading VSAM Files

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
       *> Random read (KSDS or RRDS)
       MOVE KEY-VALUE TO RECORD-KEY
       READ VSAM-FILE
           INVALID KEY
               DISPLAY "Record not found"
           NOT INVALID KEY
               *> Process record
       END-READ
       
       *> Sequential read
       READ VSAM-FILE
           AT END
               SET END-OF-FILE TO TRUE
           NOT AT END
               *> Process record
       END-READ
       
       *> Sequential read with NEXT (explicit)
       READ VSAM-FILE NEXT
           AT END
               SET END-OF-FILE TO TRUE
       END-READ

Writing VSAM Files

cobol

1
2
3
4
5
6
7
8
9
10
11
12
       *> Write new record (KSDS or RRDS)
       MOVE DATA TO RECORD-FIELDS
       WRITE RECORD-NAME
           INVALID KEY
               DISPLAY "Error: Duplicate key or invalid position"
           NOT INVALID KEY
               DISPLAY "Record written successfully"
       END-WRITE
       
       *> Append to ESDS
       MOVE DATA TO RECORD-FIELDS
       WRITE RECORD-NAME

Updating VSAM Records

cobol

1
2
3
4
5
6
7
8
9
10
11
12
13
       *> Update existing record (KSDS or RRDS)
       MOVE KEY-VALUE TO RECORD-KEY
       READ VSAM-FILE
           INVALID KEY
               DISPLAY "Record not found"
           NOT INVALID KEY
               *> Modify record data
               MOVE NEW-DATA TO RECORD-FIELDS
               REWRITE RECORD-NAME
                   INVALID KEY
                       DISPLAY "Error updating record"
               END-REWRITE
       END-READ

Deleting VSAM Records

cobol

1
2
3
4
5
6
7
8
       *> Delete record (KSDS or RRDS, not ESDS)
       MOVE KEY-VALUE TO RECORD-KEY
       DELETE VSAM-FILE RECORD
           INVALID KEY
               DISPLAY "Record not found for deletion"
           NOT INVALID KEY
               DISPLAY "Record deleted"
       END-DELETE

Comparing VSAM Organization Types

VSAM Organization Type Comparison
Aspect	KSDS	ESDS	RRDS
Access method	By key (indexed)	Sequential only	By relative record number
Record order	Sorted by key	Write order	By position
Random access	Yes (by key)	No	Yes (by position)
Sequential access	Yes (in key order)	Yes (in write order)	Yes (by position)
Updates	Yes (REWRITE)	No	Yes (REWRITE)
Deletes	Yes (DELETE)	No	Yes (DELETE)
Keys required	Yes (RECORD KEY)	No	No (uses position)
Best for	Key-based lookups, updates	Logs, audit trails	Fixed-size lookup tables

Best Practices for VSAM

Follow these best practices:

Choose appropriate organization: Use KSDS for key-based access, ESDS for logs, RRDS for position-based access
Handle file status: Always check FILE STATUS after file operations
Use INVALID KEY: Handle INVALID KEY for random access operations
Close files properly: Always close VSAM files when done
Validate keys: Ensure key values are valid before random access
Handle duplicates: Understand duplicate key behavior if allowed
Use appropriate access mode: RANDOM for direct access, SEQUENTIAL for processing all records
Test error conditions: Test with missing records, invalid keys, and file errors
Document file structure: Document key fields, record structure, and organization type
Consider performance: KSDS indexes provide fast access but require maintenance

Common VSAM Patterns

Pattern 1: Key-Based Lookup

cobol

1
2
3
4
5
6
7
8
9
       *> Look up customer by ID
       MOVE SEARCH-CUSTOMER-ID TO CUSTOMER-ID
       READ CUSTOMER-FILE
           INVALID KEY
               DISPLAY "Customer not found: " SEARCH-CUSTOMER-ID
           NOT INVALID KEY
               DISPLAY "Found: " CUSTOMER-NAME
               *> Process customer record
       END-READ

Pattern 2: Sequential Processing

cobol

1
2
3
4
5
6
7
8
9
10
       *> Process all records sequentially
       OPEN INPUT CUSTOMER-FILE
       PERFORM UNTIL END-OF-FILE
           READ CUSTOMER-FILE
               AT END SET END-OF-FILE TO TRUE
               NOT AT END
                   *> Process CUSTOMER-RECORD
           END-READ
       END-PERFORM
       CLOSE CUSTOMER-FILE

Pattern 3: Update Record

cobol

1
2
3
4
5
6
7
8
9
10
11
12
       *> Update customer balance
       MOVE CUSTOMER-ID-VALUE TO CUSTOMER-ID
       READ CUSTOMER-FILE
           INVALID KEY
               DISPLAY "Customer not found"
           NOT INVALID KEY
               ADD TRANSACTION-AMOUNT TO CUSTOMER-BALANCE
               REWRITE CUSTOMER-RECORD
                   INVALID KEY
                       DISPLAY "Error updating record"
               END-REWRITE
       END-READ

Explain Like I'm 5: VSAM

Think of VSAM files like different types of filing systems:

KSDS is like a filing cabinet with an index - you look up a name (key) in the index to find the file, and files are kept in alphabetical order
ESDS is like a diary - you write entries in order, and you read them in the same order you wrote them, but you can't go back and change old entries
RRDS is like a row of numbered boxes - you know box 5 is in position 5, box 10 is in position 10, and you can go directly to any box by its number

So VSAM is like having different ways to organize and find your files - by name (KSDS), in order written (ESDS), or by position number (RRDS)!

Practice Exercises

Complete these exercises to reinforce your understanding:

Exercise 1: KSDS Lookup

Create a program that opens a KSDS file, accepts a customer ID, and looks up the customer record. Display the customer information if found, or an error message if not found.

Exercise 2: ESDS Logging

Create a program that writes log entries to an ESDS file, then reads them back sequentially and displays all log messages.

Exercise 3: RRDS Lookup Table

Create a program that uses an RRDS file as a lookup table. Read a record from a specific position, update it, and write it back.

Exercise 4: Sequential Processing

Create a program that processes all records in a KSDS file sequentially, calculating totals or performing operations on each record.

Exercise 5: Update Operations

Create a program that reads a KSDS record, modifies specific fields, and uses REWRITE to update the record. Handle errors appropriately.

Test Your Knowledge

1. What does VSAM stand for?

Virtual Sequential Access Method
Virtual Storage Access Method
Variable Storage Access Method
Virtual System Access Method

2. Which VSAM organization provides indexed access by key?

ESDS
KSDS
RRDS
PSDS

3. Which VSAM organization stores records in write order?

KSDS
ESDS
RRDS
All of them

4. Which VSAM organization provides access by relative record number?

KSDS
ESDS
RRDS
None of them

5. What access mode is used for random access to KSDS?

SEQUENTIAL
RANDOM
DYNAMIC
INDEXED

6. What must you set before reading a KSDS record with RANDOM access?

File status
Record key
Relative record number
File pointer

COBOL Tutorial

COBOL VSAM Concepts

What is VSAM?

VSAM Organization Types

KSDS (Key Sequenced Data Set)

KSDS Characteristics

Defining KSDS in COBOL

KSDS Sequential Access

ESDS (Entry Sequenced Data Set)

ESDS Characteristics

Defining ESDS in COBOL

RRDS (Relative Record Data Set)

RRDS Characteristics

Defining RRDS in COBOL

VSAM File Operations

Opening VSAM Files

Reading VSAM Files

Writing VSAM Files

Updating VSAM Records

Deleting VSAM Records

Comparing VSAM Organization Types

Best Practices for VSAM

Common VSAM Patterns

Pattern 1: Key-Based Lookup

Pattern 2: Sequential Processing

Pattern 3: Update Record

Explain Like I'm 5: VSAM

Practice Exercises

Exercise 1: KSDS Lookup

Exercise 2: ESDS Logging

Exercise 3: RRDS Lookup Table

Exercise 4: Sequential Processing

Exercise 5: Update Operations

Test Your Knowledge

Related Concepts

Related Concepts

File Operations

File Organization

Indexed Files

Sequential Files

Related Pages