z/OS concepts
Previous topic | Next topic | Contents | Glossary | Contact z/OS | PDF


Quick reference: Data set structure

z/OS concepts

Working with data sets requires an understanding of the physical and logical structure of a data set, and how z/OS® accesses information in the data set.

Data set
In z/OS, a data set is a named collection of related data records that is stored and retrieved by an assigned name. A data set is equivalent to a file in other operating systems. Data sets are stored on tape or disks.
Direct Access Storage Device (DASD)
DASD is another name for a disk drive. Additional synonyms include: disk volume, disk pack, or Head Disk Assembly (HDA).
Space
Disk space is allocated in units called cylinders, tracks, or blocks.
Cylinder
A disk drive contains cylinders. A cylinder is a unit of storage on a count-key-data (CKD) device with a fixed number of tracks.
Track
Cylinders contain tracks, which are circular paths on the surface of a disk or diskette on which information is magnetically recorded and from which recorded information is read. Tracks are in count-key-data (CKD) format, which means that each track contains fields that indicate the start of the track and the space used, followed by records containing three fields:
  • The count field defines the length of the record
  • The key field contains optional accounting information
  • The data field contains the user data
Record
Tracks contain records. A record is some number of bytes containing data. The record is the basic unit of information used by a program running on z/OS.
  • Records have a logical record length (abbreviated as LRECL); different types of DASD impose different maximum lengths for records.
  • Records are either fixed length or variable length in a given data set. Traditional z/OS data sets have one of five record formats (abbreviated as RECFM): Fixed (F), fixed blocked (FB), variable (V), variable blocked (VB), or undefined (U).
Blocks
Records can be grouped into data blocks, which are the units of recording on disk. Blocking makes processing more efficient because z/OS can access an entire block at once instead of reading or writing records individually.

Block size (abbreviated as BLKSIZE) is the physical block size written on the disk for fixed (F) and fixed block (FB) records. For variable and undefined (V, VB, and U) records, block size is the maximum physical block size that can be used for the data set.

Extents
Space for a disk data set is assigned in primary and secondary extents. An extent is a contiguous number of disk drive tracks, cylinders, or blocks. Data sets can increase in extents as they grow. As with blocking, the use of extents is more efficient because reading or writing contiguous tracks is faster than reading or writing data that is scattered over the disk.
Volume
The term volume is often used to refer to a disk.
Volume serial
The six-character name of a disk or tape volume, such as TEST01.
Device type
A model or type of disk device, such as 3390.
Organization
The method of processing a data set, such as sequential.




Copyright IBM Corporation 1990, 2010