Basic concept of Host

TSO
TSO is the abbreviation for TIME Sharing option and is the MVS command input processor.
TSO is under the management of the operating system, to support the ISPF menu-type session system and resource management equipment (RMF), and the application and the system provides a variety of tools are ISPF support, the completion of their respective tasks.
RMF
RMF is an IBM product for performance analysis, capacity planning, and problem diagnosis.
Data Set
DATASET (which is the file of the PC). The name is no more than 44 characters long, with a maximum of 8 characters between each decimal, and cannot begin with a number.
When you assign a dataset, you must specify a data format, most commonly fixed-length and variable-length record formats.
Both fixed-length and variable-length records are chunked, so the required storage space and required input and output operations are less than non-blocking records.
Chunking refers to the storage processing of multiple data records in a single block, and any time an input-output operation occurs, the entire block is transmitted, not a single record.
The two words of the dataset organization and file format correspond closely. In general, datasets contain only business data sets (name, payroll, sales, etc.), while files can contain multiple types of data (images, audio data, video data, and so on).
For business data, the database is a relatively new option compared to datasets and files.
Traditional z/OS data sets are record-oriented. Under normal use, it does not have any byte stream files in the PC System and UNIX system.
In z/OS, there is no new row (NL) or carriage return and row padding characters to indicate the end of the record.
Sequential Data Set (PS)
In order of entry, EOF indicates the end of the file. Records cannot be deleted and can only be inserted at the end of a file. Records must be read sequentially, and for record 5, the system must first read the first 4 records.
The sequential data set is also considered a physical sequence set (PS).
Can be directly viewed and modified, can be stored data or source code, it can be easily understood as the root directory on the PC "text files."
partition Data Set (PDS)
PDS is a partitioned data set, in which you can store multiple member, commonly used as a program, JCL, load library.
The datasets are organized into sequential members (Member), each of which can have one or more members, each with a unique name, which is stored in the address directory as part of the dataset.
The address directory at the beginning of the dataset contains the entry portion of each member, and the entry for each address directory contains the member name and member at the beginning of the dataset, and the address directory entry is arranged alphabetically by the member name.
After a partitioned dataset is opened, you do not need to query the entire data set, and the user can obtain any one of the member records and can add or remove any one member record as needed.
However, only the member name in the address directory is removed, and the space to delete the record cannot be rebuilt until the system function is used to reallocate the data set.
Extended partitioned Datasets (Pdse: Partitioned Data Set Extended)
Similar to the PDS DataSet, it is composed of multiple member, the only difference being that the size of the Pdse address directory is dynamic and does not require the number of directory blocks to be specified when it is created.
Therefore, when adding or removing member, the system automatically allocates or reclaims storage space.
VSAM
The Virtual storage access Method (VSAM) is an access method that provides more complex functions (functions) than other disk access methods. VSAM saves disk records in a format that is unique and not understood by other access methods.
VSAM files cannot be displayed or edited as usual with ISPF. AMS is used to define the structure of VSAM, such as files and indexes.
VSAM can organize its data in several ways:
<ESDS>
Entering the sequential dataset (esds-entry Sequence data set), this form of VSAM is now seldom used. It saves records in the order in which they are written (Qsam is more commonly used for this type of data than ESDS).
<KSDS>
Indexed sequential Datasets (Ksds-key Sequence data set), which is the most common application format for VSAM. Each record has one or more indexes, and records can be read (or inserted) by index values.
KSDS provides random access to the data. The record can also be variable length. The index cannot be changed after it is set, but the entire record can be deleted.
<LDS>
The linear dataset (Lds-linear data set), in effect, is a byte stream dataset, which is the only byte stream dataset format in traditional MVS. Many system functions for z/OS use this format, but it is rarely used by applications.
<RRDS>
Relative record dataset (Rrds-relative record Data set), a file format similar to the Relative file, which divides the entire storage into fixed-length slots and is numbered separately,
The so-called relative data record number RRN (Relative), which reads data records through RRN, provides random reads and sequential reads.
CATALOG

The file system of the host is managed through the catalog.
If you define a dataset, it will eventually be assigned to a disk volume. The VTOC of this disk volume will increase the information of this dataset, and we need to access this dataset to specify the information for this volume when there is no catalog management.
This point class tries the Windows system now. But for the host to store a large amount of data, no management is very easy to data is chaotic, do not know which volume in which data in the end, the introduction of catalog.
Through the catalog, the system directly through the dataset of the HLQ in the catalog will know the distribution of which volumes, and then to the several volumes to find the corresponding dataset can be.
There are two kinds of catalog, one is master Catalog, storage System DataSet information and user catalog, all sys1.** are stored in master catalog, so the lookup will be directly found, The master catalog is also set up in the LOADXX parameter when the system is started.
One is User catalog, which can be defined, managed, and typically used by users to manage their own datasets. For a dataset, you can use Listcat to know which catalog it is in.
Eg:listcat entries (' Ibmuser.jcllib ').
VTOC
Vtoc is also a dataset that stores information about the data set of the volume, such as location, size, and so on. There is a record on the first track on the volume. This record points to the VTOC position of the volume.
Vtoc's position is at cylinder 0, after track 0, and ends before track 65,535.
However, VTOC can only manage this disk volume and cannot manage the entire disk data, and in z/OS systems, the system uses a directory mechanism to manage the datasets in the entire system.
A directory stores information based on the name of a dataset, which means that the dataset name must be unique. Both disk and tape datasets can be compiled into directories.
how datasets are stored
disk datasets in Z/OS systems are organized on disk volumes, known as Direct access storage devices (DASD). The name of a dataset on the same volume must be unique.
Datasets can be based on the device type (? ), volume serial number, and dataset name to locate. This is not the same as the file tree of a Unix system, where the file structure of a basic z/OS system is hierarchical, and z/OS data sets do not have equivalents similar to the path names.
(1) Direct access to the storage device (DASD) is another name for the disk drive.
(2) A disk drive is also called a disk volume, a disk package, or a Head disk collection (HDA). General use of volumes (VOLUME).
(3) The disk drive contains a cylinder (cylinders).
(4) The cylinder contains the track (Tracks).
(5) The track contains data records for CKD (Count Key data) format.
(6) A data block is an organizational unit that is recorded on disk.
data Set record format
Commonly used is the FB fixed length blocked. vb variable length blocked. U variable length for the load library.
First we explain the difference between "block" and "record". A block is a combination of logical records, a unit of reading, and a record is a logical entity.

The data set record format is described below:
(1) fixed length block record format (f–fixed), refers to a physical block on the disk is a logical record, and all the block/record size is the same.
(2) fixed-length block record format (fb-fixed Blocked) refers to some logical records are combined into a physical block. This can provide efficient space utilization and operation. This format is typically used for fixed-length records.
(3) Variable length block record format (v-variable) refers to a logical record as a physical block, because the record is longer, requiring the application to insert a 4-byte record description Word (RDW) at the beginning of the record. The length of the physical block contains the length of the record and the RDW of 4 bytes, which is rarely used.
(4) Variable length block record format (vb–variable Blocked), refers to a physical block consists of several logical records, each record has a RDW description, the physical block also has bdw described as 4 bytes.
(5) undefined format (u–undefined) The format is made up of a variable length physical record/block with no predefined structure. Although this format may be attractive to many special applications, it is usually used only for executable modules.

Blksize
The block size (blksize-block size) is the physical block size written on the disk for the F and FB records. For the V,VB and U record formats, the data set can use the maximum physical block size.
Lrecl
The logical record size (lrecl-logical record Length) is the size of the logical record (F,FB) or the maximum allowable logical record size (V,VB). The format U record has no lrecl.
Recfm
The record format (recfm-record format) has F,FB,V,VB or U, as described earlier
--END--

Basic concept of Host