Storage calculations for temporary storage data sharing
You can use the z Systems® Coupling Facility Structure Sizer tool (CFSizer) to calculate storage requirements for temporary storage list structures in a coupling facility.
A coupling facility structure contains both stored data and the information needed to manage and access that data, in a similar way to a key-sequenced data set. The data for each entry in the coupling facility is stored as a chain of fixed-size (usually 256-byte) elements, which means that the exact length for variable-length data must be stored separately. To do this, CICS® includes a length prefix in the stored data, so space calculations must allow for each entry using a whole number of elements. The amount of internal control information depends on the level of functionality and performance of the coupling facility control code for the current coupling facility level (CFLEVEL). The storage requirements can increase for a higher CFLEVEL. For more information, see Coupling facility storage management.
CFSizer is a web-based application that takes these factors into account and communicates with a coupling facility at a current CFLEVEL to calculate storage requirements. See CFSizer.
- Maximum number of queues
- The maximum number of data lists that are reserved when CICS allocates the structure list.
This value determines the maximum number of large queues that can
be stored in the structure and corresponds to the MAXQUEUES server
parameter. See List structure parameters.
A large queue is one where the total size of the data items exceeds 32K. For a small queue with multiple items that does not exceed 32K, all the queue items are stored as the data portion of the queue index entry. If the queue exceeds 32K, it is converted to a form where one item per entry is stored in a separate list in the structure and is referred to by the queue index entry.
Specify a large enough number to handle large queues, but not so large that unused preallocated list headers use a large amount of coupling facility storage. The valid range is from 1 to 999999. The default is 1000.
- Average rounded item size
- The average amount of storage required for each TS queue item.
Each item has a two-byte length prefix and is stored as one or more
256-byte elements. This value determines the entry to element ratio
that is used to calculate the required structure size. The valid range
is from 1 to 32768. The default is 256.
If all queue items are approximately the same size, calculate this value by taking the average data size, adding two, and rounding up to the next multiple of 256. The amount of element storage required is two bytes more than the data item size because of the length prefix on each item.
If queue items are different sizes, round up each size first before you take the average. For example, if half the items are 100 bytes and half are 300 bytes, round up the sizes to 256 and 512 respectively, then average them. The resulting average rounded item size is 384, which is more accurate than using the average item size of 200 and then rounding it up to 256.
- Total number of items in all queues
- The total number of entries in all the TS queues.
- Target usage percent
- The percentage of the structure space that the given total number
of items are expected to use. Specify a number in the range of 1 to
100. The default is 75. This value ensures the following:
- Free space exists for temporary expansion.
- If the initial free space is not enough, there is time to expand the structure in response to warning messages (which normally start at 80%).
- Activity to alter entry to element ratios is reduced.
- Maximum expansion percent
- The percentage that the structure can expand. If a non-zero value is specified, the maximum structure size will be greater than the initial structure size by an amount such that the total amount of data can increase by this percentage. For example, if the value 200 is specified, the initial size is enough to store the specified total number of items, and the maximum size is enough to store three times that number of items.