If VM/370 monitor data collection is active using tape, a supported tape
drive must be dedicated to the system for the duration of the monitoring. For accounting purposes, all 110 is charged to the system. VM/370 MONITOR DATA VOLUME AND OVERHEAD Use of the VM/370 Monitor usually requires that three pages be locked in
storage for the entire time the VM/370 Monitor is active; however, only
two pages are required if the single buffer option is used with only the PERFORM class of data collection enabled. This reduces by three the Dumper of page frames available for paging. This significantly affects
the performance Qf the rest of the system when there is a limited number
of page frames available for paging. PERFORM This class of data collection is activated once every 60
seconds (or as defined by the MONITOR INTERVAL command). and
records system counters relevant to performance statistics.
It is, therefore, a very low overhead data collection option. RESPONSE This class collects terminal interaction data and, because cf
the human factor, has a very low rate of occurrence relative
to processor Consequently, this class causes
negligible overhead and produces a low volume of data. SCHEDULE This class records the queue manipulation activity of the
scheduler and generates a record every time a user is added to
the eligible list, added to queue1 or queue2, or removed from
queue. The recording overhead is very low. USER This class of data collection is active once every 60 seconds
(or as defined by the MONITOR INTERVAL command). Data is
extracted from each user's VMBLOK, including the system VMBLOK. The overhead incurred is comparable with that of the
statistical data of the PERFORM class; however, it increases
with the number of users logged onto the system. INSTSIM This class of data collection can give rise to large volumes
of data because of the frequency of privileged instructions in
some virtual machines. This may incur significant overhead.
It should be activated for short periods of time and
preferably, though not necessarily, when other classes of data
collection are inactive. If the Virtual Machine Assist
feature is active for the virtual machine, the data volume
and, consequently, the CP overhead may be reduced. DASTAP This class of data collection samples device activity counts
once every 60 seconds (or as defined by the MONITOR INTERVAL command) and is a very low source of overhead, similar to the PERFORM and USER classes. SEEKS This class of data collection can give rise to large volumes
of data because every start I/O request to DASD is recorded
via a MONITOR CALL. SYSPROF This class of data collection is complementary to the SCHEDULE and DASTAP classes and results in a small amount of additional
overhead. It obtains more refined data on DASD resource
usage. Part 2. Control Program (CP) 123

First you must determine how many similar users can be run concurrently
on a given configuration before the throughput of individual users
becomes unacceptable.
Every installation should use the automatic monitoring facilities to
simplify and automate the collection of performance data. A virtual
machine should also be set up to analyze and report the collected data.
The VM/370 Performance/Monitor Analysis program (VMAP) does such a task.
For more information about the capabilities of this program and for
details about ordering it, see the publication Virtygl !gcil!iIL1IQ This program or
user-written analysis programs should be run on a daily basis to analyze
the collected data. Data reduction should preferably be run at off-peak
hours to minimize the effect on the performance of the system that is
doing data reduction. Initially, the data collected with MONITOR default options should be analyzed to establish a familiarity with the
load environment and performance profile of each virtual machine system and its effect on CP. Once a performance profile is established for each system and
associated virtual machines, the analyst should be able to detect points
of contention between processor(s) storage, I/O, and paging subsystems. Normally the spool file monitoring options should be used. However,
if large volumes of trace data are to be collected, then monitoring to
tape should be used. Tape is also useful if benchmarking is frequently
done and all of the new monitor trace and sampled data must be archived
for possible future use. The default mode of operation of the
Performance/Montior Analysis Program is to keep the condensed ACUM files
and not the raw data.
If SEEKs data is needed, a sampling technique is suggested. A simple
implementation might be to use a CMS EXEC procedure to enable SEEKs for
ten seconds every ten minutes. This would produce SEEKs data while
limiting the volume of data collected. An alternative is to create a
list of devices for which data for the SEEKs class is to be collected. CP will collect data for only those devices in the list. To create the
list, use the INCLUDE or EXCLUDE options of the MONITOR command's SEEK
operand. If data is collected for only a few devices, consider
collecting data for longer periods of time. LOAD ENVIRONMENTS OF VM/370 Two distinct uses of VM/370 can be readily identified and, consequently
some differences in criteria for acceptable performance may occur. The
system may be required to time share multiple batch-type virtual
machines with interactive machines performing minor support roles; or,
the system may be primarily required to provide good interactive
time-sharing services in the foreground, with a batch background
absorbing spare resources of real storage and processor. 124 IBM iM/370 Programmer·s Guiae