There are two options for creating the output data that will be used as input to Control Center. The output of data can be to the //AVMDATA sequential file, to a USS file (UNIX System Services), or both. USS files provide an easy way of naming files for Control Center to use directly.
The standard naming convention for input data for Control Center is as follows:
YYYYMMDDhhmmss_<A><T><P><S><target>.<suffix>
Where
YYYYMMDD is a date, e.g., 20151031
hhmmss is a time, e.g., 141500
<A> is a for ASCII data and e for EBCDIC data. The output file(s) from Acquire are written in EBCDIC on the mainframe, but the process of copying the text file (AVMDATA) with FTP will convert it to ASCII on the Windows machine running Control Center. If the file is created by Acquire and USSCOMP=Y is in effect the first character will become an e as when it is transferred to Control Center it will be done in binary and will still be EBCDIC data.
<T> represents the type of data, and is s for standard and z for compressed data
<P> represents the platform, and is always z for System z
<S> is data source, v for z/VM MONITOR data or y for Velocity Software data
<target> is an 8-digit number that uniquely identifies the system, and is predefined using the System Manager interface to Syncsort Capacity Management.
<suffix> is any even numeric value – this value is not used for anything in the z/VM environment, but must be present to satisfy Control Center’s naming convention.
Do Not Use An Odd Number For The File Suffix
For architectural reasons Control Center will ignore files with an odd suffix.
Example file names for z/VM data might be:
For MONITOR data:
20150131000000_aszv00000237.4 if ASCII
20150213001500_ezzv00010410.22 if compressed EBCDIC
For Velocity Software’s zMON data:
20150131000000_aszy00000237.4 if ASCII
20150213001500_ezzy00010410.22 if compressed EBCDIC
In z/OS it is possible to create datasets with these names using the USS component that is a standard part of all z/OS environments. Facilities like TCP/IP and FTP require a USS environment to be available.
From a z/OS point of view, USS file systems are single datasets. From within USS they appear to have data stored in the typical tree-structure of a UNIX or Linux system.
For Acquire to create these files, a control card of USSPATH= is used to point it to the directory in the USS file system where they should be created. Acquire will not create the directory if it is not present, and the user running Acquire must have sufficient privileges to read and write to the directory.