Open source cracked Big data dilemma Hadoop not the only option

Once upon a time, social networks were growing quietly and becoming an integral part of people's work and life. Facebook is a typical representative of social networking today. Facebook, the leader of social networking sites, was initially designed to facilitate communication between college dormitories and later developed into a social network of more than 900 million users and ranked first in the world. According to IDC, 1 million new links are shared every 20 minutes on Facebook and 10 million user reviews are released. Jay Parikh, vice president of infrastructure technology at Facebook, recently said the site processed more than 500TB of data per day, including 300 million of photos and other Jay Paric data. Social networking sites and other Internet applications have evolved into a data architecture that collects, analyzes, processes, and adds value.

The development of social networks in China is equally irresistible. China, which has the world's first population and the world's largest PC and smartphone shipments, has created the world's largest number of internet users and mobile internet users, making data much larger than the rest of the world. Sina Weibo's latest data show that Weibo users are averaging 60 minutes a day online, and 60% of active users are logged in by mobile terminals, with 40% of all original content from mobile terminals sharing photos. The resulting large data with high capacity, type diversification, continuous update, and bring the ultimate value to users, and so on, has not been able to deal with traditional technology.

Open source technology to crack big data dilemma

Big data not only affect people's lives in all aspects, more importantly, it brings changes to the enterprise. According to Gartner, corporate data will increase by 800% in five years, with 80% unstructured data, mostly from groups, communities, and non-business data from social networks. In the face of such a scale of data, for the prepared enterprises this is undoubtedly an information gold mine, how to convert large data into valuable information will become the future of the company's magic weapon.

The commercial value of large data is not to be underestimated. For today's businesses, data is an important strategic asset, and if you can see its business value and act quickly, it will take a chance in future competition. However, traditional relational database technology is good at dealing with structured data, but not the most unstructured data in the future enterprise. At this time, a large number of new technologies emerged.

Large data from the date of birth, showing the "open source" of the gene. Closed-source, proprietary, and overall hardware storage solutions have not yet been perfected to help enterprise users cope with the impact of this unstructured data explosion, and open source software can reduce the risks associated with large data. Open source solutions allow users to leverage hundreds of PC servers to implement large data applications in a matter of seconds, with users paying only a small portion of the cost of using resources.

A series of open source technologies, represented by Hadoop, become the main force of data storage, processing and analysis. Major IT giants aim at open source technology, have launched their own large data strategy. As one of the most watched technologies, Hadoop has gradually moved to a commercial path, offering commercial versions to businesses, just as Red Hat does for Linux.

Challenging large data Hadoop is not the only option

When it comes to big data, it's often associated with Hadoop. It's undeniable that Hadoop is a very good open source tool for handling large data. In addition to Hadoop, there are a number of outstanding technologies and tools springing up, including open source language R, NoSQL database and unstructured data storage solutions, and should become an indispensable member of the large data arena. It has to mention the June 27, 2012 global listing of Red Hat Storage Server 2.0.

Red Hat Storage Server 2.0 is an externally scalable open source storage software solution, primarily for unstructured data management, and the first file storage solution in the industry that can easily be integrated with object storage, effectively scaling up to meet the growing demand for unstructured data. It can be configured on a preset platform or in a private cloud, public cloud, or mixed cloud environment to optimize storage-intensive enterprise workloads.

Red Hat Company as the world's largest open source technology vendors, the ability to harness open source can not be belittled. In October 2011, Red Cap bought the Open-source storage software provider, Gluster, with 136 million dollars, and has since accelerated the pace of the layout of the big data market. Linux is the best platform for managing large data, and red Hat leverages its strengths to build Red Hat Storage Server 2.0 on the Red Hat Enterprise Linux operating system, where users can configure cost-effective, Scalable, high-availability storage software.

Red Hat also sees the value of Hadoop for large data. Red Hat Storage Server 2.0 can be compatible with Apache Hadoop to provide users with the latest storage options for Hadoop configuration. It can be integrated with Hadoop HDFs or as an alternative to HDFs to achieve faster file access.

In addition, Red Hat Storage Server 2.0 can also be compatible with two-way X86 servers for more than 50 major hardware vendors and supports industry-leading file access protocols, including the Common Internet File System (CIFS), the Network File System (NFS), HTTP and OpenStack Swift etc.

Summary

The impact of large data continues, reflecting not just commercial value, but disruptive changes in enterprise IT architecture and technology. As Ranga Rangachari, vice president and managing director of Red Hat company storage, says, in the next few years, open source storage solutions and large-scale X86 servers will change the storage market in the same way that Linux and large-scale X86 servers change the server market.

(Responsible editor: The good of the Legacy)