We are excited to announce that with the upcoming 6.0.15 release targeted for July 2022, we will have completed the transition of all of our masking algorithms to the new algorithm library. The completed algorithm library affords our customers a number of important improvements over our legacy algorithm technology, including performance, extensibility, security, and portability.
As we have upgraded our algorithms, we have strived to make sure we are producing the same output values as before in order to minimize re-mask events in your environments. We took this approach because we are extremely sensitive to the amount of effort required to re-mask interdependent systems in your lower environments so that your integration tests still work. We have been successful in avoiding causing a re-mask event until now.
However, with the 6.0.15 release, we have removed legacy custom algorithms (formerly mapplets) and introduced new versions of segmented masking, tokenization, and the “PHONE SL” frameworks that produce different results than previous releases. Please refer to the following list for all changes:
|
Algorithm or Framework
|
Upgrade behavior
|
6.0.15 Masking output stability compared to prior releases
|
|
Secure Lookups
|
Upgraded in place
|
Masking results the same 1
|
|
Tokenization Algorithms
|
Upgraded in place
|
Masked results will change 2
|
|
Segment Mapping Algorithms
|
Upgraded in place 3
|
Masked results will change
|
|
Legacy custom algorithms (Mapplets)
|
Upgrade blocked
|
Not applicable
|
|
PHONE SL
|
Upgraded in place to dlpx-core:Phone Unique
|
Masked results will change
|
|
Binary Lookups
|
Upgraded in place
|
Masked results the same
|
|
ZIP+4
|
Upgraded in place to dlpx-core:Repeat First Digit
|
Masked results the same
|
|
Mapping Algorithms
|
Upgraded in place
|
Masked results the same
|
|
All other algorithms
|
No change
|
Masked results the same
|
___________________________________
1 Minor whitespace changes that will not affect referential integrity
2 Users should rerun all tokenization jobs, and redistribute copies of the data post-upgrade to 6.0.15.0. Reidentification jobs in 6.0.15+ will return incorrect results when operating on tokenization job output from releases prior to 6.0.15.0. Users should use a release prior to 6.0.15.0 to re-identify data from older tokenization jobs.
3 Certain segmented mapping algorithms can not be upgraded
IMPORTANT! Before upgrading to 6.0.15 or higher, we ask that you perform the necessary analysis and planning to make your upgrade process as smooth as possible. Please be sure to follow this upgrade checklist:
- Do my masking engines have legacy custom algorithms/mapplets present?
If the answer is yes, then before proceeding any further, those mapplets need to be replaced with other algorithms built in the new framework. Mapplets are not supported in 6.0.15. Please see the following EOL notice from this past year. The following community post describes how to find mapplets on your engine(s): Legacy Mapplets EOL.
- Do my masking engines have any segmented mapping algorithms that need to be modified prior to upgrade?
Upon upgrade to 6.0.15, we will automatically replace your legacy segmented mapping and tokenization algorithms with new versions in the new framework. In very rare cases where you have defined preserve ranges that span segment boundaries, you may need to modify your segmented mapping algorithm definition for the upgrade to succeed.
- Are my masked systems interdependent? That is, do they interact with each other using common masked values such as social security numbers, credit card numbers, or other identifiers that have been masked with either a mapplet, phone sl, segmented mapping algorithm, or tokenization algorithm?
If the answer to this question is no, then you can proceed with the upgrade to 6.0.15 without re-masking any of your existing masked applications.
If the answer to that question is yes, then you will need to develop a plan to re-mask any interdependent applications upon upgrading to 6.0.15 in advance of running integrated test cases.
If you have any questions, please utilize the Delphix Community. You can respond below in the comments or create a discussion for your particular concern.
#CCE#ContinuousComplianceEngine#algorithm#custom_algorithm#2022_6.0.15