Limiting Big Data Losses with Disaster Recovery Planning

Few businesses have big-data-specific disaster recovery plans that define the way information is backed up, restored and recovered following a major outage, according to Mary Shacklett of TechRepublic.

Making Big Data Recovery a Priority

Despite big data's critical role in enterprises, database administrators suggest disaster recovery plans are not necessary; they say big data is not mission-critical data and data sets are too large and require too much recovery time to make backup worthwhile. IT managers have no choice but to make recovery a priority, according to TechRepublic, as companies continue to become more dependent on analytics for making business decisions.

When IT managers craft their big data recovery plans, they may want to consider how soon they will need to restore the data, and which data sets need to be recovered following an incident. Increased mobility, a rise in BYOD policies and common risky behavior by employees in unsecured settings makes timing a crucial aspect of a recovery plan. Though outages account for some data losses, the misuse of corporate computers and unauthorized network access and application use are usually the culprits. A study conducted by InsightExpress showed that 70 percent of IT professionals believe at least half of their companies' data losses were caused by staff usage of unauthorized programs, according to Cisco.

Cloud storage or on-site rapid storage are good options for the big data used for near-real-time analytics. Early on, stakeholders should discuss which data sets are most vital and whether they want to recover raw big data or data that has already been extracted, transformed and loaded following an outage. It is therefore paramount that IT managers practice plans to ensure they work as intended, restoring the data deemed most vulnerable.

Spend Today or Spend More Tomorrow?

Data loss has the potential to create chaos. Big data makes matters more difficult because the volumes of data are so large and their types vary immensely.

Unfortunately, midsize companies probably aren't making the best decisions if they are investing in big data collection without recovery plans. One Techaisle survey suggests small and midsize businesses are expected to spend more money than ever on big data in the next few years; global small and midsize business spending on big-data-related deployments is expected to reach $3.5 billion by 2016. Midsize businesses, which often have smaller budgets and a lack of resources, may face greater financial consequences if they experience losses without disaster recovery plans in place.

IT managers should keep in mind that firms may not need to abandon their existing data recovery framework for business continuity and information life cycles to plan for big data restoration. For many, it will come down to understanding the impact big data loss will have on their businesses.

The best approach for midsize companies with big data is to allot time to discuss recovery strategies and estimate the potential financial loss that could be avoided by implementing proper recovery plans. Midsize executives may be more inclined to pay now instead of running the risk of spending more money after an outage or loss.

Big data disaster recovery may seem overwhelming, but no longer having access to analytics due to a loss or outage may be even more overwhelming.

This post was written as part of the IBM for Midsize Business program, which provides midsize businesses with the tools, expertise and solutions they need to become engines of a smarter planet. Like us on Facebook. Follow us on Twitter.