Virtualization Technology News and Information
Predictive Analytics for IT vs. BI... What's the Difference?

Predictive Analytics for IT vs. BI... What's the Difference?

A Contributed Article by Daniel Heimlich, VP Marketing & Strategic Alliances at Netuitive

"Predictive analytics" in the context of business intelligence (BI) has been getting a lot of buzz lately as organizations are challenged to manage exponential growth of business data and understand how to use it for competitive advantage.  BI frequently applies math-based, predictive analytic approaches for data mining and web trend analysis to model forecasts about business outcomes.   One example is analysis of historical data to determine how likely a customer is to exhibit a specific behavior which can aid in optimizing marketing campaigns.  Another example is customer relationship management (CRM) where methods of predictive analysis are applied to customer data to pursue CRM objectives.

There is an interesting parallel taking place in the world of information technology (IT).  Predictive analytics for IT uses advanced, automated mathematical software to collect, analyze and correlate vast amounts of real-time IT data to forecast IT performance issues before they affect quality of service for end users.   In the case of IT, the "end user" is the customer and the focus is on ensuring the best possible performance of their mission critical applications. 

As the internet becomes pervasive, so have these mission-critical applications.  Everything from retail banking to e-commerce to online gaming means big business.  The end users' experience with these applications is more important than ever and is clearly tied to the performance of the IT infrastructure supporting it.  

At the same time, technology advancements such as virtualization created new challenges for monitoring and managing dynamic virtualized and cloud infrastructures underlying the applications.  Managing the speed and complexity of IT data being generated in virtual environments now exceeds human analysis.  This had a significant impact on an enterprise's confidence in deploying their most important applications until transformational virtualization management solutions recently became available.

Predictive analytics for IT is one of these transformational changes and is having a big impact on virtualization and application performance management (APM).  Some of the early adopters of virtualization including eight of the world's top 10 banks, several global telcos and wireless messaging giants are
now using predictive analytics for IT to forecast degradations and avoid outages for their most critical applications - many of these running in large, private cloud infrastructures.

Predictive analytic software leverages advanced statistical analysis and algorithms that automatically self-learns the behavior of an entire IT environment to forecast, identify and resolve IT performance issues.  This also enables holistic visibility required for application and performance management
across platforms in large, highly dynamic virtualized environments.  This is very good news for enterprises with expanding virtual footprints seeking to realize the full benefits of virtualization and private clouds.

And while predictive analytics has been around for years, allowing enterprises to automate manual and rules based processes for physical environments, it was not until the advent of virtualization where these adaptive, self-learning analytical approaches found their home and are now proving to excel in solving virtualization and cloud management issues such as virtual stall and virtual sprawl. 

But unlike BI, which typically involves analysis of historical data to forecast long term trends, predictive analytics for IT is focused on real-time data analysis.  This real-time capability is made possible with breakthrough "Behavior Learning" technology that analyzes and correlates real-time IT data to determine "normal" IT behavior in order to detect anomalies and forecast problems before they occur.   

Categorized by Gartner as "transformational," Behavior Learning technology analyzes and self-learns vast amounts of IT data in real-time and is at the core of the most accurate predictive analytics software solutions.  

Demand is taking off with Gartner now predicting that 40% of the Global 2000 will have deployed Behavior Learning technology by 2014, up from 10% in 2010.  One global telco reported in a Gartner case study that it is using Behavior Learning technology and predictive analytics to analyze more than a million metrics simultaneously allowing it to eliminate 3,480 hours annually in service degradation representing a business savings of $18 million

This is very good news for enterprises with expanding virtual footprints seeking to realize the full benefits of virtualization and private cloud infrastructure.   Self-learning, predictive analytics for IT - delivering visibility and automated problem diagnostics across all layers of the IT stack enabling organizations to manage mission critical-application performance confidently and proactively....finally! 


About the Author

Daniel Heimlich, VP Marketing & Strategic Alliances

As a founding member of Netuitive’s management team, Daniel oversees marketing strategy for the company’s predictive analytics software which enables IT organizations to assure the performance of critical business applications and cloud infrastructures.  Netuitive customers include 8 of the world’s 10 largest banks. Previously, Daniel served in senior marketing positions for a variety of start-ups and technology leaders, including Citrix Systems, where he helped drive revenue from $40M to $600M in six years.

Published Wednesday, May 04, 2011 5:00 AM by David Marshall
Virtualization analytics | Diozemuk - (Author's Link) - June 9, 2012 2:38 PM
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<May 2011>