2. SAMPLE PLANNING APPLICATIONS › 2.16 VAX/VMS CPU Resource Planning › 2.16.1 Usage Guidelines

2.16.1 Usage Guidelines

 
The VAX/VMS CPU Resource Planning Sample Application provides
a capacity planning database file with the database elements
you need for basic tracking, reporting, and forecasting of
the VAX/VMS CPU workload. The information for this
application is derived from either the DETAIL, DAYS, WEEKS,
or MONTHS timespan of the CA MICS VMS MONITOR Modes (DEMMOD)
file and the CA MICS VMS System Profile (DEMSPR) file. This
sample application provides information about system-wide
processor use.  Data for individual CPU IDs in the DEMMOD
file are summarized to the SYSID (system node) level.
Consequently, the VAX/VMS CPU Resource Planning Sample
Application permits your installation to track and forecast
CPU use at a high level of summarization (that is, at the
SYSID and ZONE, HOUR or DAYNAME level).

In addition to providing basic information about processor
utilization, the VAX/VMS CPU Resource Planning Sample
Application also stores data concerning direct and buffered
I/O, page faulting, and Inswaps.  You may find these elements
useful in tracking and reporting. For more information about
the data contained in the VAX/VMS files see the VAX/VMS
Analyzer Guide.

NORMALIZATION
 
Another useful feature that the VAX/VMS CPU Resource Planning
Sample Application provides is the ability to normalize
processor utilization to a base CPU.  Processor busy time is
an indicator that is commonly used in capacity planning for
reporting and forecasting.  However, as systems are saturated
and upgraded to more powerful systems, a graph of processor
busy looks more like a "saw-tooth" function than a smooth,
monotonically increasing function.  This is because more
powerful systems can process the same amount of work in less
time than their slower predecessors could, and therefore
spend less time processing.  This hides the fact that the
installation has really increased its processing capability
with the upgrade.

You can circumvent this situation by normalizing all
processor busy time to the equivalent of a base CPU.  You may
optionally request that the processor busy time and processor
percent busy elements be adjusted to a base CPU of your
specification.  To use this feature, you must supply the
following information:
 
o  The base CPU model type, or name
 
o  The base CPU Vax Units of Performance (VUP) rating

If the CPU Time Adjustment option is chosen, both the
normalized and the unnormalized CPU busy and percent
utilization elements are still stored in the database and are
available for reporting and analysis.

We recommend that you use this option to normalize workloads,
since trending and forecasting CPU utilization without
normalization is extremely difficult and often misleading.

IMPORTANT NOTE
 
You must also be certain that the VUPs ratings for your
VAX/VMS systems are stated correctly in the CA MICS SYSID
parameter.  These values are used to rate the CPU information
as it is input into the CA MICS database by the CA MICS
VAX/VMS Analyzer Option.

NORMALIZATION EXAMPLE
 
The concept of normalization can be made clearer through an
example.  Suppose that an installation installed the VAX/VMS
CPU Resource Planning Sample Application. They chose the
optional CPU time adjustment and specified a VAX 8350 as the
base CPU, with a relative performance rating of 2.3 (VUP).

These parameters, entered on the CPU Time Adjustment
Parameters control screen, are then used to normalize CPU
time values for any processor appearing in the CA MICS VMS
MONITOR Modes (DEMMOD) file to the equivalent of a VAX 8350
processor.  Assume that one of the CPUs in the DEMMOD file
for this installation is a VAX 11/780, with a relative
performance rating of 1 (VUP).  During one week, this CPU had
an average processor busy time of 35 hours in 40 hours of
operation.  We can calculate the average percent busy for the
VAX 11/780 for the week as:
 
    100*(35/40)) = 87.5%.

Given that the VAX 11/780 has a VUP rating of 1, by
convention, we have chosen to use this factor for our
normalization.  Therefore, the normalized average CPU busy
time is:
 
    35 * (1 / 2.3) = 15.2  hours

In other words, this technique estimates that a workload
requiring approximately 35 hours of CPU time on a VAX 11/780
requires approximately 15.2 CPU hours on a VAX 8350.  The
adjusted CPU percent utilization would be:
 
    100 * ( 15.2 / 40 ) = 38.0%

While normalization can calculate that a CPU runs at over
100% utilized, this type of calculation shows what the
estimated utilization would be if the workload were running
on a different type of VAX System.  That is, we would need
multiple systems of a given type if the workload exceeds the
capacity of the system.

Note first that the use of an actual, commercially available
VAX system as the base CPU is not required, and may not even
be desirable for the installation.  It would be perfectly
acceptable to specify a hypothetical system.  For example:
 
    Base CPU Model   =   MY_CPU
    CPU VUP rating   =   10

These parameters adjust all processor busy elements to the
equivalent of a hypothetical CPU having a relative
performance rating of 10 VUPs.  The fact that such a CPU may
not actually exist is not important, but rather that all CPU
time is being converted to a common unit that can be trended
and examined smoothly and consistently over time.

These parameters are not dependent on the number of
processors in the VAX system under study, since both the VUP
rating and the summarized data represent the VAX system as a
whole rather than individual processors in the CPU.

Also note that these calculations make no attempt to apply
any analytic queuing formulas or benchmark estimates to
compare one processor to another.  The basic comparison of
systems is calculated as the ratio of relative performance
ratings for the respective processors.  These numbers can be
supplied by the manufacturer or can be established by the
installation through benchmark testing.