FB "Big Data" 45-day consumption 1 million: not all companies need

Maybe you don't need big data. The 2012 "Big Data" is booming, but the author of the paper, SiSense vice president of the data analysis company Bruno Aziza, says that not everyone needs big data. "Big Data" is everywhere.　　From social media start-ups to Central Park in New York, every company seems to be deploying big data analysis. The data from Gartner, a leading data analyst, seems to be proving the point: A recent report shows that big data will drive 2012 of global IT spending of $28 billion trillion, up from $230 billion trillion by 2016.　　230 billion dollars is almost the annual gross domestic product of Portugal. However, you need to pay a large price to deploy large data technology solutions.　　Most companies do not have so many it budgets, nor can they afford data scientists or data analysis teams.　　If companies that provide large data services want to provide services to businesses of all sizes, there are several issues that need to be recognized and addressed.　　Big data is too expensive! You've probably heard of those brilliant cases that use Big data: Facebook stores about 100TB of user data a day, and NASA processes about 24TB of data a day.　　These figures are indeed impressive. So how much does it cost to process this data?　　According to Amazon's redshift pricing, NASA will need to pay more than 1 million dollars for the 45-day data storage service. According to a recent survey, CIOs in most companies say their budgets cannot afford large data deployments.　　The cost of data storage and processing is so high that we need to look for other solutions to keep smaller companies from being shut out of big data. The key to big data is not "big". Now the world's largest technology companies need to deal with petabyte-scale data.　　However, SAP research suggests that 95% of businesses typically use only 0.5TB to 40TB of data. The example of Facebook and NASA is the exception, not the norm. The fact is that processing data is not a patent for a large company. If you look at the size of an American company, you will find that there are over 50,000 companies with only 20 to 500 employees, most of which have a need to solve data problems. So the biggest demand in the big data market is not from the big Fortune 50 companies, but from Fortune 500000.　　Why do we focus on only a handful of exceptions and ignore companies that have most data-processing needs, but are neither Fortune 50 nor PB-scale? Sometimes I wonder what happens if we change the definition of big data. Often people use 3V (velocity,volume,variety) to describe big data, so let's say: "Big data is a subjective state that describes the situation when a company's infrastructure does not meet its data-processing requirements." "ItA definition may not be so glamorous, but it will certainly be closer to today's reality.