Taming Big Data: 12 Best Practices for Analysts

Now that IBM has sunk its teeth into the big data market, it wants companies to start using its technology correctly in order to get the most out of their raw information.  The vendor’s latest infographic outlines 12 best practices that users should take into account in order to realize this goal.
Big Blue places management and disaster recovery at the top of the chart: organizations must gain visibility into their infrastructure and prepare for worse-case scenarios if they want to get their big data under control. Achieving operational efficiency is an even more complicated task.
IBM lays out the prequisites for this next stage: an organization must be able to scale-up rapidly and cost-efficiently; it must be able to do the same with backup; and all applications and processes need to be optimized to meet business requirements. The Big Data leader also stresses the need for tight security, a clearly defined set of policies governing the usage of data, and the ability to effectively audit the entire stack.
Replication is number nine on the list, followed immediately by virtualization: users need reliable access to data, while admins require tools that can help them make use of all the available resources on the network.   IBM’s final two best practices are archiving, for the purpose of future analysis, and constant availability.
Big Blue’s stance is that that Hadoop, real-time analytics and leading NoSQL platforms still have a long way to go before becoming truly viable for businesses, but it has no intention of waiting it out. The company launched a new series of Hadoop-powered SMB servers just last week, a day after announcing the acquisition of Star Analytics.


Original Source