Want to work in large data, how to learn to lay the groundwork

http://www.aliyun.com/zixun/aggregation/13584.html "> Mass data is divided into two pieces, one is System construction technology, two, massive data application."

First of all, system building, now the mainstream technology is Hadoop, mainly based on MapReduce distributed framework. Now you can learn this first. But my point is that before the distributed system comes out, it's mostly a centralized architecture, like db2,oracle. Why the distributed architecture now, because the centralized architecture is limited by IO performance, slow to come out, if another hardware technology, can quickly deal with large amounts of data, performance to meet demand, then the centralized architecture is superior to the distributed architecture, because the centralized architecture is stable, the pressure of operation dimension is small. Now the centralized architecture is either not performance-required or too expensive. I look forward to a technology that can transmit and process data very quickly, so that the centralized architecture will get into people's eyes again. Again, massive data applications. Data mining and machine algorithms are the main applications of mass data. There are different application scenarios, such as personalized search and referral, social networking discovery, precision marketing, precision advertising, real-time optimal path, artificial intelligence, and so on. See if you want to do system support technology or the combination of business applications.

If you are learning system building technology now, you can read the following books:

If you learn data mining and machine algorithms, it is recommended to first look at the introduction of data mining, statistical analysis principles, Mahout,r,matlab