Fast DB2 Data Manipulation, Data Anonymization, Data Reduction
Making hi-speed copies of DB2 databases and table spaces is only half
of the story. The ability to mask or anonymize sensitive data fields ( Personally Identifiable Information PII, SSN, etc. ) and to select
data subsets from large tables is often pushed off to other in-house
applications. Most organizations quickly realize that maintaining these
home grown utilities can create a substantial workload on their DB2
resources, both staff and hardware. BCV5 addresses these issues in an
elegant and efficient manner. Following are just some of the relief
that this product provides:
- One-pass processing means Elimination of multiple jobs to copy, mask / anonymize and
- Significant reduction in hardware
utilization and total wall clock elapsed time
- Automates: scheduling, DB2 structure
object validations, copying
Select Data Fields
(Anonymization or "Data Scrambling")
normally need to copy sensitive data from a production system to a
testing or QA system. In order to adhere to data privacy regulations
and corporate security, a significant amount of extra processing is
often required. Sarbanes Oxley (SOX), HIPAA, and other compliance issues add
time and overhead... BCV5 processing and speed helps you meet your goals.
copying data like account balances or medical records to a
development environment, perhaps the actual names of customers should
not be visible to an application programmer. Data scrambling or replacing the name and
other sensitive information with random data is sometimes required.
In some cases, it is desirable to replace characters according to a
given pattern; this provides the ability to reconstruct the original
input for parallel and integration testing strategies.
return to main BCV5 product page -
copy and entire table when a smaller and more manageable subset will
get the job done? Depending on the stage of testing, a full copy of
a table may not be necessary. If a particular test scenario only
requires customers from a certain zip code area, why copy an entire
table with thousands or millions of unwanted rows?
tables grow in size, there will not always be enough free space to
hold multiple copies of all testing data. Therefore, at times it may
be preferable to specify criteria for copying or skipping rows from
large tables prior to executing the actual copying or replication.
The BCV5 solution allows the anonymization and data reduction functions
to be made in-flight. At no time does the target environment contain
sensitive or unwanted data rows. The task of coding multiple job
streams and scripts that unload data, manipulate sequential data sets
and reload the data is eliminated. There is no plethora of
statements to execute on the copied table, which results in huge DB2
logs, and numerous constraint checks and index updates.
It is possible to make a list of table columns that require
anonymization by BCV5. Every copy task containing a table on that list
is automatically processed according to the rules specified by the
organization. Individual copy tasks can also have additional rules that
only apply only to that task.
Data reduction is implemented using a WHERE clause, similar to one used
in a regular SQL statement. This facility even works when the source
table space is stopped.
Data is manipulated by a proprietary technology built into the copy
utility that operates directly on the underlying VSAM data sets. The
manipulation takes place outside of DB2 which is one factor that
accounts for the product�s speed. No log data is
produced that could
be used to draw conclusions about the original data. After the copy,
the indexes on the affected tables are automatically rebuilt.
Optionally, an image copy is taken as well. When the process finishes,
the target tables are ready for use.