During backup and restore processing, CA Disk processes VSAM data at the cluster level rather than the data space level. Logically, data processing accesses records within a file. Physically, the file is a group of areas (extents) on one or more DASD volumes. For VSAM, the cluster entity is equivalent to the logical file, while the data space is equivalent to the physical storage. Processing data at the physical (data space) level makes the data dependent on both the storage medium (device type) and the method of access.
An archive system can retain data offline for an extended period of time and when restoration of the data file becomes necessary, the storage medium can no longer be available. Also, in the event of a disaster, the recovery site cannot have the same physical medium or access method available. Processing data at the logical file level is one way of providing data independence; that is, the capability to restore the logical data to a different physical medium or possibly using an alternative method of access. In addition, logical processing allows the data to be reorganized as a by-product of processing.
CA Disk uses normal access methods to read VSAM clusters. It can read them in one of two ways:
The method CA Disk will use is controlled by sysparm VSACCESS. The basic difference in the methods is that with logical record processing, CA Disk issues requests for individual data records—it is up to VSAM to read the control intervals and perform buffer management functions. With control interval access, CA Disk requests specific control intervals that it must, in turn, deblock into their component logical records. This deblocking process can be suppressed, however, when necessary (such as when processing DB2 data bases).
With control interval access, CA Disk performs its own buffer management. The size of the buffer area is equal to the size of one data component Control Area size.
CA Disk will read all Control Intervals sequentially within a Control Area and store them before processing the logical records within the Control Intervals. This technique of reading the entire Control Area is done to take advantage of the read ahead functions of VSAM. By reading the Control Intervals sequentially, VSAM will automatically read the entire Control Area when the first read is issued. The result is a single read I/O for each Control Area processed.
|
Copyright © 2015 CA Technologies.
All rights reserved.
|
|