Will virtualization help bring order to our increasingly noisy data houses?
I recently helped conduct a survey of 381 data managers on this very question, and found a great deal of interest in virtualization as a way to better manage the overwhelming flood of data washing through enterprises. However, because virtualization requires support from multiple departments, it will take some political and organizational skills to make it happen.
The survey, conducted among members of the Independent Oracle Users Group (IOUG) as part of my work with Unisphere Research, a division of Information Today Inc., confirms that not only are data volumes and database instances on the rise, but these environments are siloed across multiple vendors, running on multiple platforms.
For instance, more than eight out of 10 respondents report that the number of database instances within or across their companies have increased over the past year. A majority of respondents have mixed database environments, and there is a limited amount of sharing of data that takes place between these environments.
Database virtualization – or the decoupling of the database layer from hardware and applications – is still a relatively new approach for many enterprises. Most already have server virtualization in place, and half are looking at database virtualization to increase the value of their data environments. However, many virtualization projects are still in early or pilot phases, and data virtualization is not widespread. About 10 percent of survey respondents could say that virtualization has reached most of their production database environments. A majority of respondents, 55 percent, intend to boost database virtualization. Once implemented, database virtualization enables users to access disparate data sources from anywhere across the enterprise, regardless of underlying database or data type.
The survey finds that the organizations with the most advanced virtualization efforts underway report that they are able to more effectively grow their data capabilities while reigning in staff time and costs. Those that are embracing high levels of virtualization are more likely to be expanding their production database environments, signifying that virtualization paves the way to such expansion.
Since data resources from many parts of the organization need to be engaged, political and leadership skills are required to make data virtualization work. Among those not employing virtualization strategies, the most prevalent reason is a lack of a business strategy, cited by 28 percent of organizations not employing virtualization. Twenty-two percent report they still are getting familiar with the technology, but a similar number report they are not happy with licensing and support issues for virtualization from their database vendors. Respondents across the board are most affected at this time by budget issues in managing their database environments. This also presents the greatest challenge to virtualization efforts at many sites as well. More than a third of respondents say budget constraints are the greatest business obstacle to their virtualization efforts.
Today’s data environments are growing more complex, with greater demand from business users for more flexibility and access. The survey demonstrated that organizations that are employing virtualization in a big way are able to more effectively grow their data capabilities while reining in staff time and costs.