What is VSAM?

VSAM stands for Virtual Storage Access Method. It is a method of managing files that is used mainly on mainframes but also on PCs. Specifically, VSAM can speed up access to file data by using a reverse index of records appended to files. This index is called a B+ tree.

VSAM is a kind of record-oriented file system. In this kind of dataset, information is stored as a collection of records. VSAM records can be of any length; they need not be of one set length. They are, however, organized into blocks called Control Intervals, which are measured in bytes. These Control Intervals are further organized into Control Areas, which are measured in much larger units.

VSAM can be organized into three types of dataset: Entry Sequenced Data Set (ESDS), Key Sequenced Data Set (KSDS), and Relative Record Data Set (RRDS). ESDS items are listed in the order in which they were input, regardless of any other consideration. Each item in a KSDS is assigned a unique numerical key, according to which the data set is indexed. RRDS items are organized by related record numbers, which can be set by users.

A perfect example of VSAM that millions of people use everyday is a FileMaker or Microsoft Access database. Accelerated spreadsheets, these databases contain a varied amount of columns, all of which contain alphanumeric information. Financial institutions have databases containing personal information on customers, and this is the kind of database we're talking about. For example, Henry Gale lives at 815 Wayland Avenue in Wayzata, Minnesota. Each item of information would be stored in its own column, for easy comparison.

These VSAM database applications allow users to configure the data in a variety of presentations and search parameters. The least search-friendly, at least initially, is ESDS, since its only method of organization is the potentially random order in which the data was entered. VSAM data that has keys or related records is much easier to present in an orderly fashion. Individual searches for names or numbers can be conducted, but more sophisticated search possibilities must have data organized in a different way in order to be most effective.