Making cabinet copies
Cabinet copies provide a performance enhancement when you are copying a large number of spaces. In such cases, the overhead to open and close each copy data set can be a significant component of the overall runtime.
Cabinet copies allow you to copy all the spaces for a specified output descriptor into a single data set called a cabinet file. The cabinet file is allocated and deallocated only once, regardless of the number of objects that are copied to or recovered from the cabinet file. Because there is no file opening or closing for each space in the cabinet file, the file header and trailer records, including the EOF markers, are omitted from cabinet files, and performance is greatly improved. You can copy the cabinet files to either DASD or tape.
In addition to providing a performance enhancement, cabinet copies can save resources because using cabinet copies can
- Save disk space because of the efficient use of space within a cabinet file
- Reduce the number of MVS catalog entries
- Reduce VTOC entries
- Reduce time for multiple recoveries or copies of image copies (COPY IMAGECOPY)
You can also use COPY IMAGECOPY to read a cabinet copy and create a non-cabinet copy.
You can use MODIFY ICFDELETE for cabinet copies, but the cabinet file will be deleted only when all members of the cabinet have been deleted from BMCXCOPY.
BMC AMI Recover will automatically use cabinet copies if they are available, for recovery of any spaces in the cabinet copy.
This section contains the following topics: