Robust IT architecture ensures data integrity

"To ensure data integrity, a company has to define a data policy, define a data classification and tiered storage strategy, and design and manage an effective archiving solution," says Shekhar Dasgupta, President and COO of Solix Technologies, in an interview with Usha Prasad of CIOL.

Excerpts:

CIOL: Before getting into specifics, is the Indian SMB community aware of the benefits of data warehousing, and subsequently, integrity of data?

Shekhar Dasgupta: The need for a Data Warehouse transcends the boundaries of geography, size, and verticals. Textile/garments exporting houses is a typical example. They need to anticipate the demands of their principles, as the time window to meet their requirements is only a season long. And each season brings orders of new designs depending on the fashion in vogue. Quite a few of them have realized that competitive advantage comes from Business Intelligence.

CIOL: How serious are the emerging enterprises in ensuring data integrity?

SD: A mid-sized manufacturing company in India was having problems in the initial days of its enterprise resource planning (ERP) implementation. It turned out that the Bill of Materials transported from their legacy systems was completely out of sync with other applications. The need for data integrity dawned on them with harsh reality. It is a company that now runs multiple applications using a common data model, and has implemented Data warehouse.

CIOL: How does data integrity help emerging enterprises in balancing performance and decent returns?

SD: In the above example, the lack of data integrity impacted manufacturing schedules that resulted in delayed orders, penalties, loss of productivity and ultimately impacted bottom line. Investments in Service Oriented Architecture (SOA), a common data model and a unified platform for enterprise data management are the cornerstones in today's world for better performance management. The return on investment (ROI) can be calculated from the opportunity cost of not doing it - the impact it has on the bottom line.

CIOL: What steps should an SMB take to preserve data integrity? How can you help them go about it?

SD: The causes of data integrity problems in high growth emerging companies need to be first examined. Typically, these companies start with a non-scalable application that meets its immediate commercial needs - and more often than not the application covers only a specific area like a financial accounting. Soon, there is a need to cover another area, say production and inventory control. The two applications have different data property rules and are usually not integrated. The problem is initially ignored, as it does not seem to make a material difference to the operations. A third area - Service management - emerges and the problem gets compounded.

Soon, the immature IT system appears to be the bottleneck as there's missing orders, wrong or delayed shipments, inaccuracies of customer data and poor customer service.

The organization then goes for an ERP. While that by itself cannot be the panacea, the first thing that the customer needs to ensure is bringing in the old data from their earlier applications into the new system and ensure that this time the data is of good quality. Otherwise, it will run into the problems.

To ensure data integrity, the company has to go through the following steps:

 Define a data policy: what data needs to be retained from the applications being sunsetted as well as going forward, for compliance and business operations purposes

 Define a data classification and tiered storage strategy: for data being retained, there are different classes depending on its changing value over time and appropriate tiers of storage need to be defined for each class to balance the need for long term data retention versus costs of maintaining such historical data, and

Design and manage an effective archiving solution: an effective archiving solution manages all data types - structured, semi-structured and unstructured - from a single platform with easy manageability and search.

Solix Technologies can help in all three areas with its unified platform - Solix Enterprise Data Management Suite that manages all data types with an integrated Google Enterprise Search Appliance - and the comprehensive range of data management services it provides along with its partners.

CIOL: What are the glitches that must be avoided so as to not to compromise data integrity?

SD: While the evolution of IT systems in line with the growth of an enterprise is inevitable, it is highly recommended that in its fairly early stage, an enterprise should embark on an IT architecture that is flexible but at the same time robust and scalable; has a data governance policy that ensures compliance to regulatory issues without making data retention burdensome; and above all lays out a comprehensive data security and privacy policy that is to be strictly adhered to by all concerned parties, including external stakeholders.

CIOL: How can you, as a vendor, help SMBs in this exercise?

SD: The heart of the Solix Enterprise Data Management Suite is a Metadata Manager - a repository, which has and can capture all relationships of multiple applications in a customer's environment. Having done that, we can help meet nearly every aspect of a customer's data management needs - data archiving, data security, data auditing, data quality, and application sunsetting.