Big data: Five things CIOs will know

1. Understanding Large Data











Large data analysis originated from large network service providers, such as Google, Yahoo, Twitter, they need not only the data generated by users, but also want to maintain the competitiveness of enterprises.





your company may be a very small company, there will be a lot of data. "In the next few years, many industries, including healthcare, the public sector, retailing and manufacturing, will benefit from big data analysis," said GigaOm research director Jo Maitland, a consultancy firm in a recent report.





collects and analyzes transaction data to give the organization a deeper understanding of customers ' preferences. It can be used to better understand the creation of new products and services and easier to respond to emergencies.





2. Useful information may appear anywhere





Hortonworks CTO Baldeschwieler says you may not agree with big data right now, but you'll recognize it, with big data using data you've discarded before.





large data can be your server's log file, for example. The server keeps track of the websites and pages that each visitor visits, and their addresses, and tracks the data to find out what the user likes. The visitor's Landing data is not novel, but it can find new information from it.





3. The lack of a large number of suitable talents





when the enterprise set up a large data analysis system, but the corresponding http://www.aliyun.com/zixun/aggregation/14294.html "> Large data talent is very scarce." James Kobielus, a Forrester research analyst, says big data relies on solid data modelling, focusing on data science, which may require statistical models, text mining, and emotional analysis, unlike the skills currently available to ordinary analysts.





Big Data talent may be in short supply, McKinsey predicts that the US could face a shortfall of 140000 to 190000 in-depth analysts and 1.5 million managers and large data analysts by the end of 2018.





In addition, enterprises need to have the ability to store and process large amounts of data, the management of 100 servers is completely different from the management of 10 servers, so need to hire a super administrator.





4. Large data does not require prior organization





CIOs typically classify various types of data, and collect into the Data warehouse, this is the first step, then based on the set of data patterns to pave the relevant data, which means that you import data before you know what you want to do, if you later change the idea, the data is limited.





Therefore, there is a large data warehouse, like "garbage dump", the first run the analysis program, and then find the relationship. Many CIOs do not know what they are looking for until they collect data.





5. Big data is more than just Hadoop





Many people talk about big data, almost all of them refer to the Hadoop data analysis platform, but the big data is not just the platform.





Monash, an analyst Curt Monash, said the Lexusnexus company has recently opened its HPCC BAE platform. MarkLogic is equipped with its own database for unstructured data, MarkLogic Server. In addition, the Splunk search engine is used to search and analyze data, such as server login information analysis.

(Responsible editor: The good of the Legacy)