Previous Topic: Analyzing the CIA ReportNext Topic: Analyzing ESDS Entry-Sequenced Data Sets


Analyzing RRDS Relative Record Data Sets

There's really not much that can be done to tune RRDS clusters. Other than checking the extent count, the only other value of interest is the dead space percentage printed in the summary section. If an inefficient control interval size is used, an inordinate amount of space would be wasted. Using the formula listed under SPACE %?DEAD, try substituting different control interval sizes to see which one produces the least dead space.

Keep in mind the type of processing that is normally used on the data set. For direct processing, it is generally more efficient to use smaller control intervals, while sequential processing is more efficient with larger control interval sizes. Even though a particular CI size cannot use DASD as efficiently as another, it can be a better choice if the data set is used frequently.