Big Data: No longer just a buzzword.

Big data means a lot to many people, but how wide is its influence? Imagine these characteristics of big data and the masters who put them together.

It is no longer just a popular word.

Big data, no matter how you define it, is highly praised and maliciously slandered. It means many things to many people: it is a kind of welfare for scientists and retail shopkeepers, and it is also an available technology to deal with a large number of privacy and security threats.

Whether it is a savior or a scam, or even a combination of the two, big data is still a hot topic for authorities, prophets, marketers and security enthusiasts. Its unofficial definition is also gradually evolving. What is that? The definition of Wikipedia has made a good start: "Any data collection is so huge and complicated that it is difficult to use data management tools or traditional data processing software."

However, when the number of data sharing devices increases geometrically, the challenge of managing a large number of different high-speed (classic 3V definition) data sets is changing. These devices are collectively referred to as the Internet of Things (IoT), including machine sensors and consumer-oriented devices (such as interconnected thermostats), light bulbs, refrigerators and wearable health monitors. IDC predicts that the IoT (Internet of Things) market will grow explosively in the next few years, from 965,438+billion installed equipment at the end of 20 13 to 28 10 billion in 2020.

Useful insight into big data can help enterprises gain many potential benefits, not only selling more products and services, but also better managing health, preventing the proliferation of counterfeit drugs, tracking terrorists and even tracking your phone records. Therefore, we know that big data itself is not good or bad. What matters is how you use it.

Ironically, no matter how great the potential of big data is in improving human experience, it is usually difficult to collect, filter, analyze and explain it to get those valuable ideas and insights. This slide examines the challenges faced by big data and its ability to cope. The hard facts will surprise you. What expectations should we have? Well, it seems that Hadoop, the leading big data platform, has a bright future. Data scientists and related big data experts should get high-paying jobs in the coming year.

Industry insiders have predicted that the popular word "big data" will gradually fade out. Herb Cunitz, president of Hortonworks, wrote in a blog in February 20 12: "After all, it all comes down to data. Big data and all predictions about this space will collapse, and analysts and all those who pay close attention to it (including many "big" suppliers) will lead to "data management".

Cunitz may have foreseen the end of "big data" prematurely, but he accurately pointed out that it all comes down to data in the end. Only change the management tools. Now, please study our slides in depth and watch some statistical analysis and research reports on display.

How much data is ignored?

According to a recent Forrester study, most companies expect that they have analyzed about 12% of the existing data. Is this good or bad? Well, these companies may have missed the insight and thinking hidden in the 88% data they ignored. Perhaps they wisely avoided the strategy of running out of resources and trying to boil seawater. Forrester believes that the lack of analytical tools and "mandatory" data islands are two reasons why companies ignore most of their own data, and the reason also includes a simple fact: it is often difficult for companies to judge which information is valuable and which information is best ignored.

Big data explosion

Crazy big data is good news for skilled workers with certain skills. According to Dice, the demand for data experts on a website for technical and engineering professionals is soaring. According to the website's April report, the number of positions of NoSQL experts increased by 54% compared with the previous year, while the number of positions of "big data talents" increased by 46%. Similar Hadoop and Python talent positions increased by 43% and 16% respectively. Of course, this is nothing compared with the recruitment notice of data security experts. According to an impressive statistic, the latter has soared by 162% in the past year.

How big is big data?

The volume of the digital world will increase from 3.2ZB today to 40ZB( 1ZB is roughly equivalent to 1 100 million TB) in just six years. Rob Bearden, CEO of Hortonworks, said in the keynote speech of Hadoop Summit 20 14 held in San Jose, California: "We are very excited to see that the amount of data around us is exploding. From now until 2020, the amount of enterprise data will increase by 50 times. Most importantly, 85% of this data comes from new data sources. " Bearden pointed out that these sources include data generated by mobile devices, social media and networked machines, which are both challenges and opportunities for enterprises around the world.