This is the number of bytes that CA Disk estimates are not usable for data storage. The value is based on both absolute dead space (such as unused bytes in the last segment of a spanned record or unused space in data sets with fixed-length records) and assumed dead space (in variable-length files, space that is available for record expansion but that is not large enough to contain a data record). Keep in mind that the estimated dead space in a control interval can be reclaimed later by a control interval split.
This is the size of the smallest record that was found in the data set.
This is the size of the largest record that was found in the data set.
This is the average size of each record in the data set. For fixed-length files, MIN, MAX and AVG record sizes will be the same value.
This value is the minimum number of records that were found in any used control interval.
This value is the maximum number of records that were found in any used control interval.
This value is the average number of records that were found in each used control interval.
This is the minimum number of unused bytes found in any used control interval. This space can actually be considered dead space in other calculations of this report if CA Disk determines the space cannot be used to hold data records. Control intervals that contain a segment of a spanned record are ignored in all four of the freespace per CI values.
This is the maximum number of unused bytes found in any used control interval.
This is the average number of unused bytes found in any used control interval.
This is the average percentage of freespace to use space in all control intervals read by CA Disk.
This value represents the average number of bytes required for control information for each data record. This information is required by VSAM to access the records but is not part of the user's data. This value can be useful in estimating storage requirements when reorganizing a cluster (add this value to average record size to come up with an accurate value of space required to hold each data record).
This value represents the maximum relative deviation between any two control intervals when processed in key sequence.
This value represents the average relative deviation between control intervals when processed in key sequence. A value of 1.0 represents the best case of each control interval in key sequence physically residing next to one another. As this number increases in size, so does the disorder of the data set.
A separate print line will be printed for all clusters that have spanned records. Because of the poor performance associated with spanned records — both in CPU and DASD utilization — these values should be checked closely. If a fairly large percentage of records are spanned, you should consider increasing the control interval size of the data set (this cannot be an option in an IMS or CICS-type environment).
This is the total number of spanned records that reside in the data set.
This is the percentage of records in the data set that do span control intervals.
This is the size of the smallest spanned record in the data set.
This is the size of the largest spanned record in the data set.
This is the average size of all spanned records in the data set.
This is the amount of dead space in the data set attributable to spanned records. In essence, this is the total number of bytes that are not being used in the last control interval of each spanned record.
The summary information for the index component of the data set is for the sequence set only. Since CA Disk processes the cluster in key sequence, it does not need any of the records in the index set, which is used to access the cluster in keyed direct access mode.
This is the total number of control intervals used to hold the index's sequence set. This also equates to the number of control areas being used in the data set, since each sequence set record controls a single data control area.
This is the number of control intervals in the sequence set that have no free control interval pointers to the data component. In other words, this is the number of sequence set records that point to control areas in which all of the data control intervals are being used.
This is the total number of data control intervals that are currently — or are projected to be in the future — unreferenceable because the index's control interval size is too small. For more information, see Unreferenceable Data Control Intervals.
This is the number of control intervals in the index component that are currently unused.
This is the minimum amount of free (unused) space found in any sequence set control interval.
This is the maximum amount of free (unused) space found in any sequence set control interval.
This is the average amount of free (unused) space found in all sequence set control intervals.
This is the average number of bytes to which each key in the sequence set compressed. This value will give an idea as to how effective the compression algorithm is. It is of most value when looking at what control interval size to use for the index. If the keys are compressing well and the MINIMUM FSPC/CI value is sufficiently high, you can want to consider reorganizing the cluster and specifying a smaller control interval size for the index.
This value is similar to the AVERAGE COMP KEYL, except it includes the overhead information required to store keys in the sequence set. Therefore, if you subtract the AVERAGE COMP KEYL from this value, you will have the average number of bytes of control information overhead that is required for each index entry.
This value represents the maximum relative deviation that occurred between two consecutive data control areas in key sequence. In a completely organized cluster with no control area splits, this value will be 1. See the description of CA DEV in the graphic analysis section for a more detailed explanation of this value.
This is the average relative deviation of all control areas in the data component. Again, the ideal value is 1.0. As the number gets higher, greater seek distances are required to process the cluster in key-sequenced mode. This value is not relevant to direct key retrieval, since the sequence set is not used. Also, if the data set is on a high activity volume, this value can be meaningless, since head movement will probably occur between control area reads to satisfy requests for other data on the pack.
|
Copyright © 2015 CA Technologies.
All rights reserved.
|
|