3 Suggestions for Big data

Are your businesses getting ready to take big data programs? Either invest in the Hadoop platform or use cloud services to tap into social media streaming? Whatever your goal, we should make it easier to use large data rather than showing great ambition at the outset.

This is the experience of GCE's http://www.aliyun.com/zixun/aggregation/35685.html "> chief Marketing Officer and the big data disseminator of Marina Lucas, whose company is a cloud service provider, Also involved in many government projects. GCE Large data and analysis cloud is a scalable platform for storing and managing data, as well as a provider of user-friendly data query and analysis tools. GCE's data/cloud Platform is the framework of the U.S. General Services Administration (GSA), a government website that includes many details such as contracts, subsidies, direct loans, payments, and how the federal government spends taxpayers ' money.

In a telephone interview, Lucas said: "We are trying to turn the meaningful information we are talking about into something that is really useful to everyone." "To build a large data platform for the enterprise is a lot of pressure, so he gave the following suggestions to the enterprise, hoping to ease the anxiety of enterprises."

1. Start from scratch

Companies that are accustomed to using relational databases may not want to try large data platforms like Hadoop. "But habits are not always the best solution," says Lucas. You must take the first step. Dig the block and analyze what you can do with it. ”

For example, a retail business may need 3 months of sales data, you put it in one of these tools, and then analyze what you learned from it.

2. Avoid fear

Change is scary, especially when its costs can be costly and potentially a professional killer. Lucas said: "It could be this way--some people would say, ' My goodness, what is it? I have to change everything to get it '? ”

I think not. A large data platform can coexist with traditional relational databases. "For now, these are free things," says Lucas. You use your trading system to carry out your business, but at the same time you have new ways to get new data. ”

3. Find the Balance

"Most people have found that cloud models can coexist with their original systems, and they have achieved a healthy balance of development," said Lucas. ”

This pattern also applies to large data. "People don't know what to do," says Lucas. They worry that one system will replace another. Relative to Yu Yun, people are more concerned about the large amount of data that has brought them too much change. ”

However, the attractiveness of large data is different from that of cloud computing. The main focus of cloud computing is cost savings. The former has a wider range of reasons, including the city. The promotion of marketing and sales, and the unique value of large data-it can handle unstructured information that traditional data management platforms cannot handle effectively.

"The valuable data that companies have are locked in business intelligence tools and transactional databases for a long time," says Lucas. ”

There may be a lot of people using big data this year, but that doesn't mean there are a lot of skilled data scientists who stop training for information. Lucas said: "I'm sure data scientists are very popular with some people, especially those who can do in-depth analysis." But most of the time people don't need these deep analysis, just need to solve the data problems they are concerned about. ”

(Responsible editor: The good of the Legacy)