Subscribe

Data management sets tough landscape

Alex Kayle
By Alex Kayle, Senior portals journalist
Johannesburg, 01 Dec 2010

Thousands of leaked US intelligence documents released by whistle-blower Web site WikiLeaks might not have happened with better data protection.

This is according to Deloitte senior consultant Avinth Rabbipal, speaking during the ITWeb Data Warehousing 2010 event held at Gallagher Estate in Midrand, yesterday.

Rabbipal noted that information management strategy has become a crucial component for both business and governments.

“Business faces more change than ever before due to the economic climate and governance regulations. There is increasing reliance on information to understand change and implement faster decision-making.

“In addition, cost pressures are driving adoption of lower cost solutions such as software-as-a-service (SaaS), open source and cloud-based data warehousing solutions.”

Business disruption

According to Rabbipal, more than half of organisations will replace their primary data warehouse between now and 2015. However, he said this is going to disrupt business, as organisations build new data warehouse architecture.

“The essential requirements of a next-generation data warehouse platform will see real-time data warehousing, cloud computing, SaaS, open source software, in-memory processing and 64-bit computing.”

Data management features very strongly as a key component to next-generation data warehouses, said Rabbipal. “Master data management is one of the highest priorities for data warehousing, as data quality remains a key focus.”

He added that driving a cross-functional and consistent view of data and implementing rationalised data-related technologies are challenges that need to be addressed by a comprehensive data strategy.

“The biggest challenges with data warehouses are time to value and achieving shareholder buy-in. While an incremental approach seems to be the standard, increments are mostly ill-aligned with limited or no tangible benefits at the end of each increment.”

As a result, he said, business tends to lose interest in the project and look for the quick wins.

Skills wanting

Rabbipal pointed out that a business analyst on an IT project who has predominately focused on delivering IT projects might not be suited to deliver a data warehousing project. “Data warehousing is a different animal and needs to be treated as such.

“Data warehouse architecture skills are seriously lacking in SA. Too many data warehouses are implemented in-house with limited knowledge and experience, often delivering sub-optimal results.”

Adding to the problem, Rabbipal indicated that too many service providers deliver solutions without sufficient skills and experience, while software vendors claim to sell miracle solutions.

“There are insufficient business skills in data warehousing because there's a difficulty to understand and communicate the business needs and drivers.”

Share