Data Masking in Informatica

Data Masking is one of the latest features introduced  by Informatica and is big advantage considering the security of sensitive data stored in Datawarehouse.For eg

• SSN number of a customer
• Credit Card number
• Phone number etc…

This is done by masking the original data using various masking techniques   i.e. converting the original data into some other readable format. This is done without affecting the granularity of the table ,all the integrity constraints are kept the same.The objective behind using data masking transformation is to restrict the outside world from accessing the Customer Information.

Since data ,masking is an integrated option available in @Informatica Power centre it greatly reduces the Development effort.

Why we need data masking?

Normally in projects we have non production environments like development, testing and training . Often the IT companies are careless about the security  in the developments phase. While moving the code into actual production environment they rarely notice that the information can be leaked out.So we need to be careful here.In order to avoid this situation we can do masking at the development phase itself and then outsource the data.That we will be more safer.

Another advantage is that we can avoid the use of Dummy values in the testing and development.Instead of using Dummy values we can use the actual and then mask it and the successfully promote to the Live Environment.

Now we will go through the different algorithms used for data masking

• Nondeterministic randomization:Replace a sensitive field with a randomly generated .This can be some either any number or date field.

• Blurring: Add a random variance to the original value .Variance can be small or larger. For example adding 10 percent increase in the salary of employee.

• Repeatable masking: Maintain referential integrity by using values that are unique and different.Eg Replace the SSN number 1/45/89 with 12/22/44

• Substitution:  Randomly substitute original values with entirely different but real valuesE.g., substitute “Wayne Smith” with “Don Kelly”)

 Will keep updating……