The control area (CA) size for striped VSAM data is a factor of the stripe count. A VSAM striped data set can be striped up to a count of 16. The minimum size for an allocation is a single track. Though the maximum CA size for a non-striped VSAM is a cylinder, such is not necessarily the case for a striped data set. Traditionally that would have meant that the maximum CA size, based on 3390 geometry, would be 15 tracks. That changes with striped VSAM data sets in that the maximum CA size now has to accommodate the maximum stripe count (16), and so the maximum CA now becomes 16 tracks. Further complicating the issue is the fact that CA sizes must now be compatible with extended addressable volumes. Consequently, the CA size divided by the stripe count must be divisible into the multi-cylinder unit of 21 cylinders (315 tracks) -- 1, 3, 5, 7, 9 or 15 tracks.

For another example, take the specification of TRACKS(16 4): for non-striped, traditional VSAM, a control area size of 16 tracks is not allowed, so the allocation of primary and secondary are adjusted upward to 20 tracks with a secondary of 5, and a CA size of 5 tracks per CA -- the CA size being compatible with multi-cylinder units. For a striped data set which is also non-spanned, with a stripe count of 3, the CA size is dictated by the smaller of the primary and secondary allocation. The system would arrive at a CA size of 4 in this instance. However that CA size is not divisible by the number of stripes, so it is adjusted upwards to 6 tracks per CA. Because there are 3 stripes, the allocation amount per volume is 2 tracks. But since 2 tracks is not compatible with extended addressable volumes, it must be adjusted downward. The net result is a CA size of 3 tracks . The primary allocation must be a multiple of the CA size, so it is adjusted up to 18, and the secondary allocation also must be adjusted to a multiple of the CA size. The primary allocation would be 18 tracks, and the secondary allocation would be increased to 6 tracks.

A larger CA results in a larger number of control intervals (CIs) in the CA, resulting in a larger number of entries to index in a data set organization containing an index (KSDS and VRRDS), with the end result being a requirement for a larger index CI size.

The calculations are different for a striped data set that allows spanned records. The CA size for such a data set does not have to be evenly divisible into the multi cylinder unit of 21 cylinders (315 tracks). Because of this, a striped data set that allows spanned records is always allocated in the track managed space of each volume at define time. However, if such a data set does have a computed CA size that is EAV compatible (1, 3, 5, 7, 9 or 15 tracks), then it can extend into the cylinder managed portion of EAV volumes.

In case of a three-striped dataset with is spanned, the CA size would be first dictated by the smaller of the primary and secondary allocation. We would arrive at a CA size of 4 in this instance. However, that CA size is not divisible by the number of stripes, so it is adjusted upwards to 6 tracks per CA. Because there are 3 stripes, the allocation amount per volume is 2 tracks. Even though 2 is not compatible with extended addressable volumes, for striped data sets with spanned records, we do not adjust it downward. The CA size remains at 6. The primary allocation must be a multiple of the CA size, so it is adjusted to 18, and the secondary allocation would be increased to 6 tracks. This data set is allocated in the track managed space of the volume at define time.