Leveraging Automated Data Validation to Reduce Software Development Timelines and Enhance Test Coverage
1 like3,054 views
The document discusses the implementation of an automated data validation solution, DatatestPro, to enhance software development efficiency and test coverage while addressing challenges posed by heterogeneous data sources. By streamlining data validation processes, the solution significantly reduces manual effort, lowers costs, and improves overall quality for enterprises across various industries. A successful case at a leading insurance provider demonstrated a reduction in validation efforts by 75% and improved test coverage by 80%, leading to faster time-to-market.
Leveraging Automated Data Validation to Reduce Software Development Timelines and Enhance Test Coverage
- 1. Leveraging Automated Data Validation to Reduce Software Development Timelines and Enhance Test Coverage By industrializing data validation, QA organizations can accelerate time- to-market and improve the quality and quantity of data analyzed, thus boosting credibility with end users and advancing business transformation initiatives. Our enterprise data testing framework overcomes the impediments of traditional test automation solutions and provides the ability to test end- to-end data. This solution is intuitive and custom- izable based on data testing needs. This white paper illustrates how our new dataTestPro solution can be applied across industries to enhance the data validation process. It also elucidates how this solution significantly reduced manual efforts, and improved QA effectiveness, when deployed at a leading insur- ance organization. Challenges in Data Validation: An Industry Perspective Homogeneity in test data is a thing of the past. In fact, heterogeneity of data is now the norm across all industries. A decade ago, a data pool of 10 million records was considered to be large. Today, the volume of data stored by enterprises is often in the range of petabyte or even exabyte. The reasons: Executive Summary Quality teams across enterprises expend significant effort comparing and validating data across multitier databases, legacy systems, data warehouses, etc. This activity is error-prone, cumbersome and often results in defect leakage due to heterogeneous, complex and voluminous data. It also requires testers with advanced data- base skills. Whenever time-consuming manual data valida- tion impacts project schedules, testing efforts are squeezed to get the project back on schedule. Where each and every bit of data is of paramount importance, this is not an acceptable solution since it puts the business at significant risk. Automating data validation is an optimal solution to address these challenges. In our view, this requires quality assurance (QA) organizations to industrialize the data validation process. Therefore, we have created an enter- prise solution that streamlines the data validation process and assists in the identification of errors that typically emerge across the IT landscape. cognizant 20-20 insights | july 2013 • Cognizant 20-20 Insights
- 2. cognizant 20-20 insights 2 spreadsheets and CSV and XML files, as well as flat files and columns and rows from multiple database vendors’ software. After extraction, the transformations of the ETL process need to be verified. This can be scripted, but the effort is often time-consuming and cumber- some. • Increased testing timelines: Test cycles take longer time to complete, if performed manually, particularly when testing large pools of data. • Test scheduling: Testing teams need to execute data comparison tests after ETL processing, where the tests are executed at a specified time. For this process, manual execution is induced by a trigger. This affects the scheduling process. Quality engineers working on data warehousing projects are continuously scouting for ways to automate the ETL testing process, but with limited success. The reason: The difficulty in standardiz- ing processes across technologies, complex archi- tectures and multi-layered designs. In addition, automation tools are expensive and require an up-front investment and extensive learning curve, which prolongs time-to-value. An Integrated Approach to Data Validation To address the aforementioned demands of data validation, we have created a comprehensive solution that can be used in a variety of data testing areas such as data warehousing, data migration testing and database testing. Our solution, dataTestPro, facilitates the QA process in enterprise information environments and significantly reduces the manual effort, delivering reduced costs for various “data testing” needs by provisioning for and managing automated data validations. As an industry-agnostic approach, dataTest- Pro has been successfully implemented at large engagements in the banking, brokerage, insurance and retail industries, and is deliver- ing immense business benefits to our clients (see sidebar, next page). Eliminating Data Quality Issues Our experience with clients shows that organi- zations can eliminate data quality issues by per- forming the following actions: • An increase in mergers and acquisitions, which results in tremendous data redundancy and vast pools of data requiring validation of com- plex extract, transform and load (ETL) logic. • A greater need for data center migrations. • An increased management focus on data and data-driven decision-making related to business intelligence (BI) initiatives. While there is an abundance of hetero- geneous data, there is also a high prob- ability of inherent errors associated with erroneous data, excessive duplication, missing values or conflicting definitions. These errors lead to cost overruns, missed dead- lines and, most important, a negative impact on the credibility of the data pro- vider. Industry research suggests that only three out of 10 organizations view their data to be reliable.1 Data-related problems result in an average loss of roughly $5 million annually; in fact, it is estimated that about 20% of these companies experience losses in excess of $20 million annually.2 However disconcerting these losses are, the intriguing and often unanswered question is: “How are these losses accounted for?” Having an appropriate answer to this question would improve business operations and aid strate- gic decision-making. For data warehousing and data migration initiatives, data validation plays a vital role in ensuring overall operational effectiveness. Challenges in Data Validation Data testing is substantially different from con- ventional testing. Quality teams across organiza- tions expend a great deal of effort in making com- parisons on huge volumes of data. This entails: • Identifying data quality issues: Most organi- zations validate far less than 10% of their data through the use of sampling. This means at least 90% of their data is untested. Since bad data likely exists in all databases, improving testing coverage is vital. • Heterogeneous data: Source data is usually extracted from various sources, including Excel Industry research suggests that only three out of 10 organizations view their data to be reliable.
- 3. cognizant 20-20 insights 3 • Streamlining and accelerating data validation. • Providing an intuitive, comprehensive and integrated workbench. • Ensuring a faster and high-quality test execu- tion cycle. • Preventing data anomalies by early detection of defects in the testing lifecycle. • Facilitating extensive reuse of test compo- nents, reducing time-to-market and simplify- ing the test management process. • Increasing test coverage. different databases, such as Oracle, Microsoft and any other JDBC-compliant database. Increasing Testing Speed Regression testing of data validation is auto- mated by our solution, thereby solving the problem of “the need for speed.” Based on our experience at multiple engagements, data com- parisons and reporting is 80 times faster than a manual process. Test Scheduling Comprehensive automation – from scheduling tests through execution, data comparison and reporting – is provided in our solution, thereby increasing the speed of testing, as well as reduc- ing the cycle time and workload of the tester. The solution provides the tester with the option of scheduling tests to run at a specified time. Execution Steps Our solution automates data validation as follows: • Test case creation: Test cases for various data comparison and validation scenarios can be created using a data mapping exercise. This solution also provides the user an option to create test suites and execute multiple test cases in a single framework execution. • Execute and report: Upon completion of the test cycle, informative summary and detailed reports are generated. The user interface is simple; for example, QA analysts from a non- technical background can configure tests to operate in different modes for different types of comparisons. Test cases for various data comparison and validation scenarios can be created using a data mapping exercise. Quick Take dataTestPro in Action Our solution was implemented at a U.S.-based leading insurance and annuities provider. The client faced data validation challenges due to huge data volumes and insufficient test coverage. A typical data transaction involved 1.5 GB of transactional data and over 1,200 ETL transactions. The client was unable to implement standard tools available in the market due to the amount of data originating from hetero- geneous data sources. We implemented dataTestPro in two different projects. With this solution, we were able to reduce data validation efforts by 75% and improve test coverage by 80%. Overall, we reduced time-to-market from 18 person months to five person months. As a result of all these benefits conferred, the client was able to reduce time-to-market, enhance test coverage and improve productivity with significant reductions in cost and maintenance efforts. dataTestPro validates com- plex data transformations, providing full coverage of data and thereby doing the following: • Performing 95% of the data validation process, with enhanced coverage and reduced risk. • Providing insightful reports highlighting detailed data differences, down to the individ- ual character level. • Automating the entire testing process, from scheduling to execution to reporting. Comparing Heterogeneous Data dataTestPro compares data between different files and databases after data migration or recon- ciliation. Source data is typically pulled from vari- ous sources such as Excel, CSV, XML, flat file types (any type of delimited file or fixed-width files) and
- 4. cognizant 20-20 insights 4 Figure 1 depicts our solution’s data validation automation solution layers. Figure 2 highlights the features and potential benefits of our solution. The Bottom Line Data loaded into an organization’s data warehouse/databases comes from multiple sources and dimensions. The analysis extracted from this data supports an organization’s strategic planning, decision-making and perfor- mance monitoring systems. Hence, the expense of defects not caught early in the QA process is compounded by the additional business cost of using incorrect data to make mission-critical business or strategic planning decisions. As a result, QA organizations need to test enterprise data from multiple standpoints: data completeness, data transformation, quality, scal- ability, integration and user acceptance or confidence. These tests are essential for the successful implementa- tion or enhancement of an organization’s data. Our automated data validation solution enables enterprises to not only improve operational efficiency, but also enhance the quality of data, The return on the investment comes from the ability to measure and track data integrity over time with relatively low per-cycle overhead. Data Validation Solution Layers Figure 1 Data Source Target Layer Data Source Configuration Operator Rule Configuration HTML Excel Reports Extraction Engine Manager Test Execution Manager Test Reports Database Fixed Length File XLS File Delimited File XML File Database Schema Data Extraction Engine Cleansed Transformed Data Data Migrated in Intermediate Form Data Cleansing Data Validator Data Transformation Engine Comparison Operator Rule Source Target Mapping Single Field Rule Multi Field Operator Conditional Operator Test Results Generation Test Management Tool Conditional Comparison rules Comparison Engine Solution Features and Value Additions Figure 2 • Supports all JDBC-compliant databases and file types. • Schema comparisons. • Supports huge volume of data. • Supports command line execution of test cases. • Test case scheduler. • Summary detailed reporting. • Reuse of test cases and SQL scripts for regression. • Export/import test cases across installations. • Capture the status of test cases that are executed through command line, scheduler and through tool. • User-friendly reporting in formats like HTML, Excel etc. Core Features Value Additions
- 5. About Cognizant Cognizant (NASDAQ: CTSH) is a leading provider of information technology, consulting, and business process outsourcing services, dedicated to helping the world’s leading companies build stronger businesses. Headquartered in Teaneck, New Jersey (U.S.), Cognizant combines a passion for client satisfaction, technology innovation, deep industry and business process expertise, and a global, collaborative workforce that embodies the future of work. With over 50 delivery centers worldwide and approximately 162,700 employees as of March 31, 2013, Cognizant is a member of the NASDAQ-100, the SP 500, the Forbes Global 2000, and the Fortune 500 and is ranked among the top performing and fastest growing companies in the world. Visit us online at www.cognizant.com for more information. World Headquarters 500 Frank W. Burr Blvd. Teaneck, NJ 07666 USA Phone: +1 201 801 0233 Fax: +1 201 801 0243 Toll Free: +1 888 937 3277 Email: [email protected] European Headquarters 1 Kingdom Street Paddington Central London W2 6BD Phone: +44 (0) 207 297 7600 Fax: +44 (0) 207 121 0102 Email: [email protected] India Operations Headquarters #5/535, Old Mahabalipuram Road Okkiyam Pettai, Thoraipakkam Chennai, 600 096 India Phone: +91 (0) 44 4209 6000 Fax: +91 (0) 44 4209 6060 Email: [email protected] © Copyright 2013, Cognizant. All rights reserved. No part of this document may be reproduced, stored in a retrieval system, transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the express written permission from Cognizant. The information contained herein is subject to change without notice. All other trademarks mentioned herein are the property of their respective owners. About the Authors Balasubramaniam AV is a key member of Cognizant’s Data Testing Center of Excellence RD team and has 11-plus years of experience architecting solutions across the data warehousing, business intelli- gence and data migration project spectrum for industry leaders in the finance, hospitality and logistics sectors. Balasubramaniam has provided consulting solutions for numerous complex quality engineering and assurance projects. He can be reached at [email protected]. Karthick Siva Subramanian is a key member of Cognizant’s Data Testing Center of Excellence RD team, with operational experience spanning banking, insurance and retail sector engagements. Karthick has worked on numerous engagements focused on data test automation innovation, tools and frameworks. He can be reached at [email protected]. Balaji Suresh is a key member of Cognizant’s Data Testing Center of Excellence RD team, with progressive experience in various facets of software testing and quality assurance in the telecommu- nications, retail and banking industries. Balaji is responsible for quality assurance of Cognizant’s data testing framework. He can be reached at [email protected]. which aids in decision-making. The success- ful implementation of this requires an orga- nization-level commitment to enhance the quality of data for decision-making. While this may involve an initial investment, in the long run the cost per cycle for executing data integrity tests is significantly lowered, as was experienced by leading organizations across industries. The return on the investment comes from the abil- ity to measure and track data integrity over time with relatively low per-cycle overhead. This boosts the credibility of the QA organization due to the business’s new-found ability to use measured data for decision-making. As data validation is not a “one-time” or ad hoc process, our dataTestPro solution makes it a thoughtful, well-planned and continuous valida- tion process supported by organizational commit- ment to ensuring error-free data. Footnotes 1 ”Delivering Trusted Information for Big Data and Data Warehousing: A Foundation for More Effective Decision-Making,” Ventana Research, 2012. 2 “The Business Case for Data Quality: A White Paper by Bloor Research,” by Philip Howard, March 2012.