Hadoop and its Little friends

&http://www.aliyun.com/zixun/aggregation/37954.html ">nbsp;

Large data is not only large, but also very complex, content and form are diverse, does not conform to the traditional database on the storage structure requirements. A very important area for large data analysis is to develop database tools that can handle highly diverse workloads.

This demand also explains why tech media are so focused on Hadoop. But a recent study found that the new database field is not just a hadoop, many small and medium-sized enterprises and other data analysis users are working to develop other analytical tools, and has a relatively wide range of applications.

Unstructured data requires a variety of tool support

InfoWorld's Antony Savvas recently wrote that the IDG survey found that while the market for Hadoop was widespread, companies did not consider it a panacea. Instead, businesses typically combine them with other large data analysis resources. The survey found that 32% of the surveyed companies used Hadoop, and another 31% will deploy Hadoop over the next year. At the same time, a number of respondents are using other large data analysis tools, almost 39% of users deployed NoSQL, and another 36% are using MPP databases.

Traditional structured databases are also unwilling to be discarded. Many analysis tools include both structured and unstructured data. Traditional structured databases are also trying to add new functionality.

Big challenge, big ecosystem

For SME it, this report gives a lot of resources to deal with large data analysis. The unstructured data used for analysis comes in a variety of sources, from smart devices to data, to company running data, to consumer social media data.

For businesses, they can assemble open source Hadoop with other tools to meet demand. Alternatively, companies can choose cloud vendors to provide analysis results, and they can focus more on the company's own business operations. No matter what the business needs, an increasingly rich ecosystem of large data analysis can support them continuously.