2. PLANNING › 2.2 Supported MVS Data Structures › 2.2.6 DFHSM Migration Control Data Set (MCDS)
2.2.6 DFHSM Migration Control Data Set (MCDS)
VCC can optionally collect basic accounting and capacity
planning information about data sets that have been migrated
from standard VTOCs and moved to compressed locations in the
hierarchy of storage managed by DFHSM. Locator information
about migrated data sets is stored in the DFHSM Migration
Control Data Set (MCDS). VCC serves as a "driver" for the
collection of data from the MCDS. The actual data collection
from DFHSM data structures is performed by a DFHSM-supplied
utility program called ARCUTIL that was delivered in DFHSM
Version 2.5 to protect user-written utilities from changes in
the record layouts in the MCDS and BCDS. VCC "drives"
ARCUTIL by ATTACHing it as a subtask and intercepting its
output to squeeze out padding blanks and to add important
time-stamp and control information.
Data in a DFHSM environment can be conceptualized as
pyramidal in shape. This pyramid is meant to convey a
hierarchy of storage ranging from PRIMARY or migration level
0 to migration levels 1 and 2. For simplicity, we abbreviate
"migration level" to "ML" and thus, it is common practice to
describe data as residing on primary, ML1, or ML2 storage.
Primary is the normal location of data directly accessible by
applications. When a data set has not been used for some
specified period, HSM migrates it to ML1 storage volumes,
where the data is usually stored in a compacted form and is
not directly accessible. It can be recalled quickly to
primary storage when needed. Finally, when installation
thresholds have been reached, HSM will migrate data from ML1
DASD to ML2 storage. ML2 storage is usually a densely-packed
tape cartridge.
/\
/ \
/ \
/ \
/ \
/ Primary \
/____________\
/ \
/ \
/ ML1 \
/ \
/______________________\
/ \
/ \
/ ML2 \
/ \
/________________________________\
In a typical DFHSM environment, the amount of data migrated
through the above hierarchy could be presented as the above
pyramid. For example, if 400 gigabytes of primary storage is
available for active data, then perhaps a much larger
quantity of data is inactive and has been migrated to less
expensive levels of the hierarchy.
The VCC driver for DFHSM data collection anticipates a large
quantity of inactive data and is designed for optimum
execution time performance and optimum data compression of
the output file it creates.
One of the principal benefits of integrating DFHSM data
collection into VCC and subsequently into CA MICS is that
there is no longer any need for capacity planners or
accounting specialists to view DFHSM-controlled data as a
special case. For example, using VCC in conjunction with
CA MICS products, you can establish different pricing for
each level of the storage hierarchy. Using the pyramid
again, you could conceive a rate structure that reflects the
relative cost of each layer, as shown in the following
example:
/\
/ \
/ $ \
/ 0.10 \
/ per \
/ Mb/hr \
/____________\
/ \
/ $0.05 \
/ per \
/ megabyte hour \
/______________________\
/ $0.002 \
/ per \
/ megabyte hour \
/ \
/________________________________\
