The data component analysis graph provides a picture of what each used control area in the data component looks like. Each control area is represented by a separate box graph. At the conclusion of each control area, summary totals are printed for the appropriate fields in the control area.
This graph contains one print line for each control interval in the control area. Most graph lines will consist of four different print characters:
A legend is printed under the final line of the graph to indicate the use of these characters. In control areas that contain spanned records, you will also see the following symbols:
Represents the control interval containing the first segment of a spanned record.
Represents any intermediate segments.
Represents the last segment of a spanned record, followed by *'s to indicate the dead space in the last control interval.
The legend for spanned records will only appear after those control areas that actually contain at least one spanned record. You can also notice some control intervals with a literal of — FREE — inside them. These are control intervals that are marked as free in the sequence set CI and therefore contain no data.
We will now take a closer look at each of these values.
The characters (D) are used to represent the space actually used to contain data records. The number of characters in all of the data records in this control interval can be found under the column BYTE TOTAL.
This is space that is currently not being used in the control interval and is therefore available either to add new records or to expand existing ones in variable length files. This space is represented by spaces.
This is the amount of free space in a control interval that cannot be used to add other records, so CA Disk categorizes it as dead space. The character * represents dead space. In fixed-length record files this amount is a constant value and is calculated as follows:
CISIZE - (RECSIZE x RECORDS PER CI) - 10
For variable-length files CA Disk makes an estimation of dead space based upon the records processed to that point. Any free space quantity that is less than the average record size +3 is considered dead space. The +3 is added in to assume the worst case of not more than two consecutive records having the same length records, which would cause three bytes of control information to be added to the control interval. It should be noted that this space is available to expand existing records. Also, even though the space is considered dead now, a control interval split later on would make the area usable again.
The unused space in the last segment of every spanned record is also marked as dead space. It is a VSAM restriction that any control interval that contains a portion of a spanned record cannot contain any other records, even if there is enough free space to hold additional records.
|
Copyright © 2015 CA Technologies.
All rights reserved.
|
|