Gajus - Fotolia
Once a practice largely reserved for fortunetellers and clairvoyants, predicting the future has become a valuable...
tool for disaster recovery planning.
As predictive analytics technologies mature and prove their worth in sales, healthcare and other fields, a growing number of organizations are beginning to realize that the technology can also be used to make disaster recovery (DR) plans more accurate and perceptive. Predictive analytics and AI are powerful disaster recovery planning tools in IT's ongoing evolution.
"Until recently, the automation of IT has largely focused on standardizing certain repetitive tasks or computations and automating them, something that's not easy for humans to do manually," said Oussama El-Hilali, vice president of products at data backup service provider Arcserve. "With AI and machine learning, combined with predictive analytics, these systems are not only capable of thinking and acting like humans, but also have the ability to foresee and forecast events in the future with a certain level of accuracy."
Predictive analytics can help DR planners model disaster events, costs, recovery, impacts and other essential strategy elements to provide greater insight than simply guessing or relying on a hunch as to what might happen, said I-Sah Hsieh, program manager for corporate social innovation at analytics services provider SAS. "By combining historical data from previous emergency events with data from sensors, applications, experts, statistics and more, today's emergency managers can produce better models to mitigate risk, reduce costs, improve resilience, better identify vulnerabilities and save lives," he explained.
Common failure scenarios can be determined by using machine language models and studying the data collected from sensors, machines or systems responsible for running the infrastructure, said Jason Cutrer, founder and CEO of Six Nines IT, a cloud services consulting firm. "Insights can be collected and analyzed to turn what was once a manual process into an automated and highly resilient system."
The first step in adding predictive analytics to your slate of disaster recovery planning tools is identifying the key questions the technology can help answer. Once these questions have been identified, look for the data that promises to generate the most accurate results. "Data is the fuel that makes all analysis work," Hsieh said.
Start with the data you already have. "Predictive analytics can help you get the most value out of that existing asset or answer some new questions you've never asked before," Hsieh said. "Sometimes, the simple act of combining data sets from different systems that have not previously met can open up a flood of new insights."
Integrating predictive analytics results into emergency communications, power backups, travel, secondary locations and other disaster recovery planning tools is essential to a successful initiative. Otherwise, one is simply adding more complexity to an existing strategy. "Understanding what's where, who is affected and what is affected gives you the opportunity to get ahead of or respond more quickly to a situation," said Mike Orosz, senior director of threat services and technology transformation at enterprise software and service provider Citrix Systems.
Getting started early is a vital aspect of integrating these disaster recovery planning tools into your recovery strategy. "The insights we get from machine learning and analytics enable us to move faster and deal with unexpected situations," Orosz said. "The sooner you can get in front of it, the sooner you can identify potential gaps in your plan to ensure the continuity of your business and the safety of your people."
While it may still be considered a novel technology, predictive analytics offers benefits that could make it one of the more popular disaster recovery planning tools in the near future. "Applying predictive analytics before disaster strikes should be a core part of every disaster recovery plan, as it provides important information before locations are in crisis," Hsieh said.