Big data rise based on cloud

Big Data is the industry's future trend, its momentum has been unstoppable, and Hadoop as a larger distributed computing and storage offline processing cluster representatives, in this year is more red all over the world.

Admittedly, big data is becoming an important factor in changing people's lives, and the integration of big data and cloud computing is in full swing.

From the quantitative, structural world, to the uncertain, unstructured world. This transformation enables us to understand the real information, improve the level of decision-making, and when the community has a more complete and analytical capability for natural data, our ability to grasp and predict events will increase. Cloud-based information storage, sharing and mining tools provide a tool for knowledge production, which makes decisions more accurate by analyzing and predicting large data, and for many businesses, data is becoming a decisive factor in winning a business battle.

Recently concluded the 11th session of the Dalian Soft Rendezvous "2013 Data Innovation Forum", with large data analysis and service as the theme, invited senior experts from Intel, Microsoft and IBM to share the views of large data technology, to explore the construction of large data ecosystems.

China has a huge population and application market, high complexity, full of changes, such a large group of users, constitute the world's most complex, most complex data. To solve this problem caused by large-scale data and to explore the solution based on large data is an important means to upgrade and improve the efficiency of China's industry.

IBM "Wisdom computing" leads the way

As a highlight of the forum, IBM China Development Center chief Technology Officer Mao in the "Wisdom of Computing" theme of the speech: Big data is not Hadoop, Hadoop is only a small branch of large data analysis, large data analysis is a very diverse area, Each field has its own mathematical model and mathematical attributes, and these models and attributes require developers to develop domain-related algorithms, and then map the algorithm to a mechanism that can be distributed on a large scale, often using the cloud infrastructure.

As a leader in the Big data field, IBM is leveraging leading methodologies and comprehensive data technology to help companies rethink their existing it models, and to leverage business transitions based on the information revolution to gain competitive opportunities and immeasurable commercial value. At the same time, IBM as the industry's only ability to set consulting, services, software and hardware comprehensive strength, providing end-to-end comprehensive integration of solutions vendors, related products and services covering large data and analysis related areas of various stages, various fields; IBM's big data practice has gone deep into all sectors of the global marketplace, including China, Including telecommunications, finance, medical, retail, manufacturing and many other customers have been in the IBM large data platform and large data analysis technology and the concept of support to obtain benefits. In addition, IBM has launched a portfolio of results-oriented solutions, including fraud prediction, financial operations and customer service three core competencies. This solution is a broad collection of IBM's intellectual capital, including software products, infrastructure, research and consulting services.

IBM launched a variety of products and technologies for large data developers, greatly enhancing the accelerated performance of its analytics products and simplifying development and operations. The new version of the enterprise-class Hadoop product Infosphere biginsights can leverage existing SQL technologies to more easily develop application components. Both security and high availability features, the new version of Biginsights can be more fit with the enterprise deployment requirements. The new stream computing software Infosphere streams further enhances the simplicity of operation in the development and deployment of real-time analysis. The new relational database management software Informix contains timeseries acceleration, which can be applied to the operational reports and analysis of intelligent instrumentation and sensor data.

Intel soft and hard synergy, experience first

"Big data is profoundly changing the software industry, it industry, and all aspects of people's lives," he said. "The industry is being reconstructed," said Dr He Jingxiang, general manager of Intel Asia Pacific Research and Development Co., director of Intel's Software and services division of China. We hope to share with the vast number of industry chain partners and developers of Intel's large data technology platform and computing solutions, to promote the construction of the local ' end-to-end ' large data environment, together to meet the increasingly demanding data challenges in the construction of intelligent cities and towns. ”

At present, Intel's software support has been extended to the entire software stack, from platform hardware, firmware, virtual machines, cloud computing to the operating system, development tools, middleware and applications, to help a wide range of industry chain partners and developers to release innovative potential, innovative user experience. With Intel's open architecture core product line for large data, with Intel Hadoop distribution, you can achieve "soft and hard synergy, experience first" innovative results. For example, using the Intel Xeon processor platform for network and I/O technology optimizations, and a strong combination of Intel Hadoop distribution, the previous analysis of 1TB data need 4 hours long wait to complete processing, now only a short 7 minutes to complete, greatly enhance the large data "intelligence" The meaning.

Intel Hadoop now has a wide range of applications, including smart cities, telecommunications, finance, healthcare and other industries that generate users. When it comes to specific cases, Dr. Miaokai points out that China has taken a big step ahead of the rest of the world in the application of Intel's Hadoop release software. In the telecoms industry, for example, the Intel Hadoop release has been used by China Unicom (600050, shares) to query and analyze historical billing and traffic. It is because of the strong customer demand that the Intel Hadoop, which was originally an innovative and tentative project at Intel, eventually turned into a real product.

By working with local partners, Intel is applying large data technology to innovation in smart medicine. Intel Asia Pacific Research and Development Co., Ltd. Data Center software Department large data Solution manager Wangxiaodong describes the collaboration that has been achieved in the past year with Ufida under the UEP framework, primarily based on the Intel x86 cluster +x86 integrated solution, to try to address current medical problems. For example, if a doctor inquires a patient case, it may have to be a case before, and as the whole medical system upgrades in the large data age, all medical cases are integrated on the Intel large data platform, so that its query speed, processing speed to reach the top level. Over the past year, both sides have experimented with varying degrees in regional, large electronic health records and regional medical quality analysis.

From it hardware systems to data management applications, in the field of data processing and analysis software, Intel has a wide range of partners and healthy ecosystems that, through close collaboration among partners, has been able to provide users with servers, storage devices, and network devices that apply to large data applications or are further optimized for this application , as well as data management and data analysis software that can be seamlessly docking with these devices and more in line with the needs of large data applications.

The reliance of large data-processing tasks on an open architecture platform has given Intel more and more developer attention in this area. As the core component of the world's leading open architecture platform, the processor platform provider, Intel's primary mission is to build a "core" for large data applications, and then help developers build the infrastructure for large data applications around this foundation. Intel also plans on its data organization and management layer, provides software-optimized Hadoop products and services for the distribution and management of large data, and provides support for client and server-side algorithm development in the analysis and discovery layer to meet the performance and scale requirements for large data calculations , the decision support and IT services layer will combine partners in the ecosystem to provide a more streamlined visual application experience. All Intel's specific policies related to these plans will be deepened and refined as large data technologies, applications and markets evolve.

Microsoft Walks "cloud" with Oracle

Combined with the recognition of large data itself and Microsoft's own powerful product portfolio, Microsoft can provide users with an end-to-end large data solution. In short, is the backend to the next generation of SQL Server 2012 as the benchmark platform, the large data tube, and then in the middle of the data mart, with a wealth of data applications, and finally in the front-end with rich interface form to show the results of data analysis, complete data--application and analysis-- The complete process of rendering the results.

To help businesses quickly adopt their large data solutions, Microsoft will also provide Hadoop on Windows Azure Platform and Windows Server (used as a cloud-based service on the latter as a distribution of internal deployments). And there are a lot of third-party tools for Hadoop that can be better analyzed and visualized on this platform. Microsoft's large data solution also supports custom mining algorithms in C + +, C #, Python, Ruby, and Pearl languages through Hadoop streaming, supported by generic third-party tools and frameworks such as Mahout.

Last September, Microsoft reached a consensus with China Telecom's sky-wing platform, which built its services and disaster preparedness services on Windows Server, and consolidated it into a very complete "cloud" operating system. For customers, this system will bring considerable value, it can put all of your applications, all the operational dimension of the scene online and offline better management. From a holistic perspective, the cloud's management system can help IT staff solve a range of operational scenarios.

To allow more start-ups to use Microsoft's Public cloud service, Microsoft Greater China Vice President and marketing strategy general manager, Microsoft Greater China Chief Cloud computing strategy officer Shenwey said that the company has launched the Cloud Accelerator project in China, free of charge for start-ups to provide Windows Azure services, The purpose is to further promote public cloud services. Windows Azure's Open, reliable, easy-to-use platform service dramatically lowers start-up system development and operating costs, helps start-ups operate quickly around the world, and delivers a consistent, user experience for global users, enabling start-ups to focus on core business development and accelerate business growth. The second phase of the project is nearing completion.

In addition, the good news that attracts the attention of developers in HTML 5 Web apps, Apache cordova/phonegap and Windows Phone 7.5 is now available with Azure as a service support. The mobile-side browsers that can enjoy this right include IE8 and "Current versions of Chrome, Firefox, Safari and more than 2.3 phonegap." ”

Recently, foreign media reported that Oracle and Microsoft announced in the field of cloud computing in-depth cooperation. Microsoft Cloud Platform Windows Azure will fully support Oracle databases, Java, WebLogic, and Linux. By working with Oracle, Microsoft can offer more deployment options to customers of Windows Azure platform. But it is unclear how the Azure platform will behave in Oracle and Microsoft's cloud.

The Big three eight recount

2012, from the entire cloud computing solution Itbrand List, IBM, Microsoft, Intel, although the ranking of slight fluctuations, but has been the top three.

Cloud computing technology is the most important and effective means to solve large data problem sets at present. Cloud Computing provides an infrastructure platform for large data applications to run on this platform. Big Data is the industry's future trend, its momentum has been unstoppable, and Hadoop as a larger distributed computing and storage offline processing cluster representatives, in this year is more red all over the world. Intel, on the one hand, has prepared a comprehensive end-to-end, core hardware support spanning a single, dual, and multiple Xeon processor product, and the Intel Hadoop distribution for Apache Hadoop, led by the Intel China team, provides powerful technical support not available in open source versions , and optimized for the Intel architecture hardware platform; IBM's strengths lie in its information management platform, information consolidation and governance, analytics, and balanced-configuration IT infrastructure; Microsoft wants to integrate Hadoop into Windows systems, Windows Azure, Windows Hadoop is fully supported on server, SQL Server 2012.

In the area of large data analysis, while the lack of technical staff temporarily, but more and more people began to learn Hadoop, enterprises are also encouraging and training technicians to learn Hadoop. The vast number of developers should seize large data opportunities, choose more suitable platform technology, with the best solution, using large data to develop a more intelligent, more personalized new generation of applications, and ultimately realize the transformation of the application economy upgrade.