Open source Hadoop develops fast users and vendors

At the heart of large data, Hadoop is an open source architecture for efficient storage and processing of large data. Open source start-ups Cloudera and Hortonworks have been in the market for years, with Oracle, Microsoft and others wanting to take a place in the market, But more indirectly, by partnering with professional Hadoop start-ups, to compete in the marketplace.

Large data core (image source Google)

According to the latest report from Forrester Analysis, traditional technology suppliers will launch a series of strong product strategies, even though there are still many areas to be perfected in the Hadoop products. But manufacturers want to provide stable products based on open source projects, directly through the cooperation of open source technology to obtain product results, but because of the contribution in the field of open source too little, it is difficult to have too much influence.

Hadoop: not selection but development

It is easy to grasp the development trend of Hadoop, as Forrester says, the future development of Hadoop depends largely on the changing data infrastructure, and Hadoop has become "the cornerstone of future flexible data management platforms". For technology suppliers, it only needs a report on Hadoop to keep up with the pace of today's business.

Companies are interested in the talk of Hadoop because it enables businesses to store and analyze large amounts of data at less cost. According to Forrester, companies currently analyze data that accounts for only 12% of the company's total data, because companies do not know how to handle data.

Hadoop can easily implement large data because it enables businesses to store data at lower cost, and once they find the best way to analyze the data, they can process the data.

Even so, some companies view Hadoop as a "unsupervised digital dump", with the knowledge of Hadoop and large data being increasingly understood, the enterprise combines Hadoop analytics with the NoSQL real-time data processing engine to collect valuable parts from enterprise data and take action in a timely manner.

What kind of supplier should be concerned about the future development of the Hadoop field?

But Forrester suggests we not just focus on the suppliers who invest a lot of money in Hadoop.

In the area of open source, it is even more important to provide the source of code than to occupy it. Similarly, proprietary software vendors charge for software licensing because it is secretive about their intellectual property, and the market power of open source suppliers is directly linked to the influence of suppliers on open source projects, especially intellectual property.

There are some different points of view about Forrester's analysis. "Instead of focusing only on the current state of the vendor Hadoop service, Forrester suggests that the real winners in the future should be large technology companies such as IBM, AWS and pivotal," said Matt Asay, vice president of MongoDB. and Hadoop startups make the biggest contribution to Hadoop development.

Surprisingly, while the strategy includes licensing, pricing, execution, product roadmap, and customer support, Forrester does not comment on all corporate community outreach activities. As ReadWrite reported earlier, with little contribution to Hadoop development, vendors like IBM and AWS find it hard to determine the direction of Hadoop development, and Hadoop founder Doug Cutting asserts:

I don't expect all companies to contribute a project to the community, but those who do not contribute will have no say in the field of Hadoop.

Similarly, Hortonworks chief executive Rob Bearden said: "The community is the key to ensuring a sophisticated link between Hadoop technology."

Hadoop is becoming the core of the data architecture, and we are still trying to make a contribution to the community and make sure that the new Hadoop kernel is able to play its potential on the next generation of data platforms.

The area of open source is not entirely fair. Amazon, for example, spends a lot of money on MySQL, more than MySQL or Oracle. Similarly, IBM spends more money on Linux than the Linux leader, Red Hat.

But Forrester's views may not be entirely correct, for example, in the area of "customer Support", where Hortonworks's contribution can be scored, because Hortonworks has contributed a lot of code to the Hadoop community, but pivotal and IBM have scored the corresponding points, This does not seem persuasive.

Forrester Analytics finds that IBM and AWS provide better customer support for Hadoop proprietary products. But those who do not contribute much to the development of open source projects, their customer support capabilities can not be compared with a lot of suppliers. While IBM, Amazon, and other companies will build quality services based on Hadoop in the future, it is hard to provide good customer support because they are always forced to be followers of Hadoop, not leaders.

The strong in Hadoop technology

The Forrester vendor rankings for Hadoop are not related to Hadoop itself, more about how Hadoop fits into the company's long-term product strategy, such as IBM's product strategy, from the product roadmap to the continued integration of IBM's other products, such as the SPSS advanced analysis, Load management, BI tools, and data management and modeling tools for High-performance computing are integrated with biginsights Hadoop solutions.

Hadoop is a good addition rather than a major part of these vendors.

Companies such as IBM, AWS and others who enrich their products through Hadoop can benefit from the market, but for those who want to gain value from Hadoop itself, it may be easier to derive benefits from companies that contribute more to Hadoop development (Cloudera and Hortonworks), and provide better support for Hadoop. I think that IBM, AWS and other manufacturers to use open source Hadoop technology to improve the product promotion market is not a problem, but to provide users with the best technical support for the Hadoop community to promote long-term and development is the long-term plan.