Dataset

A dataset refers to a larger, coherent dataset that depending on the computer and development environment has different meanings: A ' dataset ' is despite linguistic similarity not to be confused with a ' record ', but is either ( in IBM jargon ) identical to file or contains (with. NET etc. ) purpose- in columns and rows aggregated information from possibly several database tables.

Dataset ( IBM mainframe )

This dataset is another name for a file that exists on an IBM mainframe system. A dataset name (DSN ) may be up to 44 characters long and consists of multiple qualifiers ( part names ) that are separated by periods. Each qualifier can be up to eight characters long. Example: MEIN.PRIVATES.TEST.DATASET.V1

In application programs is typically accessed directly through the dataset name on a dataset. Instead, access via a logical name occurs (even data definition (DD) name called ), which refers to a corresponding DD statement of a job that contains the DSN and optional additional information for processing.

Datasets can exist in different file organization forms:

  • Sequential files: the data is sequentially written to or read from file beginning to the end of file This file form.
  • Direct Access dataset: The relative address of a data set is calculated using a hash function from the key.
  • Partitioned data set (PDS and PDSE ): A file organization in which the dataset contains a directory of member names, each member represents a single sequential file again.
  • VSAM organizational forms: The operating system component VSAM provides different forms of organization are available, whose powerful, KSDS (key - sequenced dataset ), supports the key-based access to individual records. VSAM organized files are also called VSAM cluster; to the individual VSAM forms of organization, see the article VSAM.
  • ISAM dataset ( Indexed Sequential Access Method): Deprecated form of organization that has been virtually replaced completely by VSAM. An ISAM dataset consists of three physical files ( PRIME, and INDEX OVERFLOW ).
  • HFS dataset: A plate portion is provided for receiving Unix files ( in Unix terminology, one would say: a loopback file system).

The term of the dataset is used only for files on the mainframe, which were created under the MVS Personality. Files (that is, within an HFS dataset ) created under the Unix System Services are not usually referred to as datasets.

The metadata ( file attributes) that are part of the dataset in the VTOC (for disk drives ) or tape label ( magnetic tapes ), partly in the Catalog.

Sequential files can be versioned as a Generation Data Groups ( GDG ).

Dataset (. NET)

In the. NET development environment from Microsoft a dataset called a managed in main memory related data set in tabular structure. Typically, the contents are read out with a SQL query from possibly several tables of a relational database and ' connected ' ( join), alternatively requested by a Web service, then edited locally on a PC, on a web server or application server and, if necessary, stored back into the database. Synonym in some other development environments: Recordset.

In the Service Oriented Architecture ( SOA) data changes in the dataset are summarized in a Diffgram and this returned to the data providing Web services.

So a first dataset usually contains a copy of the data from a particular ' view' on the database, but also serves as the intermediate storage of new or changed data. It is the central data buffer in aspx or smart - client applications.

Comments

  • Data structure
Row (database) Table (database) .NET Framework
219609
de