Error! Unknown document property name.
Error! Unknown document property name. E R R O R ! U N K N O W N D O C U M E N T P R O P E R T Y N A M E . E R R O R ! U N K N O W N
DOCUMENT PROPERTY NAME. Error! Unknown document property name.
Test Data Management
1/7
Error! Unknown document property name.
Error! Unknown document property name.
Error! Unknown document property name. E R R O R ! U N K N O W N D O C U M E N T P R O P E R T Y N A M E . E R R O R ! U N K N O W N
DOCUMENT PROPERTY NAME. Error! Unknown document property name.
1. Overview:
In Software development lifecycle, a significant amount of time is assigned to Testing and in testing a
good amount of testing efforts are consumed in creation and management of Test Data. The efforts
included are in form of time and money which can overall impact the delivery of product. This paper
throws light on management of Test Data using CA’s Data Manager .
2. What is test data management?
Test data is any information that is given to system as an input for execution of test case. Test Data
management is the process of fulfilling data needs for execution of test case ensuring that the data
provided is in right quantity, correct environment and is provided at appropriate time. It makes sure
that the provisioned data provides has all flavors of data, is referentially intact and is in right volume.
Data provisioned must not be too large as production data and at the same time it should not be too
small to cater all the data requirement. The data can be provisioned either by creation of synthetic
data or by extraction of production data or by sourcing from different database lookup tables.
TDM can be implemented efficiently using well defined processes, manual methods and proprietary
tools. It can also be put in use using full-fledged TDM tools such as Data Maker, Optima.
3. Challenges in TDM :
1) Test data coverage can be incomplete and testing team might not have the knowledge required
to create complete data set.
2) Data requirements and it’s volume specification is not gathered in requirement gathering phase.
3) Lack of access to upstream and downstream systems.
4) Availability of production data in huge volume that is too much to handle for test environment.
Also this data can be sensitive in nature and might be unsuitable for the test scenarios to be tested.
5) Data can get corrupted when used by different testing teams in same environment.
6) Requirement of huge data volume by testing team in a short span of time and lack of required tools
with test team.
7) Large amount of time is spent on data gathering rather than in test data creation.
4. Steps/Strategies in TDM:
1) Analysis of test Data:
Test data is generally constructed based on the test cases to be executed. End to end scenarios
must be identified, based on which the test data in created. The data required can be scattered
and a thorough analysis of data requirement in required for better data management.
2) Data setup to mimic production data:
Error! Unknown document property name.
Error! Unknown document property name.
Error! Unknown document property name. E R R O R ! U N K N O W N D O C U M E N T P R O P E R T Y N A M E . E R R O R ! U N K N O W N
DOCUMENT PROPERTY NAME. Error! Unknown document property name.
An extension to step 1 and involves development of understanding about how end user will use the
system. Data is created and modified based on the production scenario to minimize defect slippage.
3) Test Data clean up:
In scenarios where test cycle span is longer, we might require Test data cleanup at some point of
time in a test cycle. Hence a clear process of data cleansing activities should be incorporated.
4) Data masking:
In order to properly test application, sometimes we need large amount of sensitive data. This data
should be protected because of regulatory acts or privacy controls. Mechanism to shield sensitive
data from getting revealed should be identified.
5) Automation:
Test Data management includes scripting, masking of data, data cloning and data provision.
Automation of these activities can decrease test cycle time and can make the process efficient.
During Data management process, test data is linked to different test cases and can be given as
input to some automation tool. Automating this process ensures the quality of test data during
test cycles.
Similar to regression testing, production of test data should be automated. It helps in replication
of huge production type data and saves time in longer runs and helps to expose errors in ongoing
process. QA team would be in better position to streamline and validate TDM efforts .
TDM tools:
Informatica Test Data Management tool
CA Test Data Manager (Datamaker)
Compuware’s Test Data Management
InfoSphere Optim Test Data Management
HP Test Data Management
Error! Unknown document property name.
Error! Unknown document property name.
Error! Unknown document property name. E R R O R ! U N K N O W N D O C U M E N T P R O P E R T Y N A M E . E R R O R ! U N K N O W N
DOCUMENT PROPERTY NAME. Error! Unknown document property name.
CA Test Data Manager ( Datamaker)
CA test data manager is a tool that provides synthetic data generation solutions. This product of CA
technologies, is flexible and simplify the functionality of testing. This solution automates one of the most
time-consuming and resource-intensive problems in Continuous Delivery: the creating, maintaining and
provisioning of the test data needed to rigorously test evolving applications. The suite uniquely combines
elements of data subsetting, masking, synthetic, cloning and on-demand data generation to enable
testing teams to meet the agile testing needs of their organization. It has a test data repository where
generated data is stored for future reuse.
Key functionalities:
Provides in depth profiling from multiple databases and data sources
Creates test data from production data by providing tools that profile, desensitize, and slice the data
while maintaining its referential value
Stores test data scenarios as reusable assets in a central repository
Analyzes and assesses current environment and identifies data gaps to fill for complete data coverage
Improves test quality by increasing test coverage through intelligent data generation
Allows to share and reuse data through a test data warehouse known as TDM Portal
Error! Unknown document property name.
Error! Unknown document property name.
Error! Unknown document property name. E R R O R ! U N K N O W N D O C U M E N T P R O P E R T Y N A M E . E R R O R ! U N K N O W N
DOCUMENT PROPERTY NAME. Error! Unknown document property name.
Components of CA Test Data Manager:
1) Data Profiling:
Data Profiling is the process of analyzing existing data and applying filters to
discover potential personally identifiable information (PII). CA’s Test data manager
provides a good solution to identify any Personally Identifiable Information (PII)
across heterogeneous data source. Once this profiling is done efficiently we can
make business decision to archive, secure, encrypt or delete the PII data.
The different persona-based flow for Data Profiling is as follows:
Benefits of Profiling
Gain a better understanding of data structures through improvements to the existing
model.
Reduce the time taken to understand the model.
Trace through complex data models and structures to discover all independent
entities.
Improve the design of data models to ensure they meet all requirements.
Error! Unknown document property name.
Error! Unknown document property name.
Error! Unknown document property name. E R R O R ! U N K N O W N D O C U M E N T P R O P E R T Y N A M E . E R R O R ! U N K N O W N
DOCUMENT PROPERTY NAME. Error! Unknown document property name.
2) Subset Production data:
Data subset involves creation of a smaller referentially correct copy of a larger
database with huge data. Once sub setting is done, the cut-down database
remains perfectly usable for testing. The data is referentially consistent and
referentially correct but in manageable volumes and without impacting
environment’s performance.
CA TDM Data subset uses native database utilities and makes sure that highest
possible performance is achieved while extracting small, more intelligent subset
from production environment. This allows us to quickly provide manageable sets of
consistent production like data. It also minimizes the risk of exposing sensitive
records.
Benefits of Sub setting
Reduce infrastructure costs on storing and maintaining data
Greater control on testing needs
Designing exactly the data needed
3) Data Masking:
One of the major challenge we face while taking a copy of production data is that of
Data security. Production data might have sensitive information like SSN, Account
details, Card details etc. There may be some government norms on certain type of
data and client may face a penalty if that data gets compromised. To safeguard the
sensitive data we perform data masking.
CA Test data manager fast data masker is a simple, high performance masking tool
that helps in protection of huge volume of complex data. High performance masking
definitions can be created, saved and reused. These definitions helps us to comply
with data protection regulations. Various data masking functions are available which
helps in systematic , repeatable protection of Personally identifiable information.
Defined as rules, these functions helps to mask and anonymize data. Example of
masking rules available are:
1) Hashing
2) Credit card
3) Seed Table
4) Numeric variance
5) Random Ranges and text
6) Substitutions
Error! Unknown document property name.
Error! Unknown document property name.
Error! Unknown document property name. E R R O R ! U N K N O W N D O C U M E N T P R O P E R T Y N A M E . E R R O R ! U N K N O W N
DOCUMENT PROPERTY NAME. Error! Unknown document property name.
7) Offset dates
