Hadoop architecture replaces Data Warehouse cloud computing virtual technology to enhance storage capabilities

The traditional business intelligence system only serves for the decision person, and is affected by the security factor, the analysis data can only be reflected in the local user terminal, but manifests the way almost is the ordinary computer. With the rapid development of computing and Internet technology, the mobility and activity space of sales, market and decision makers are more and more, more and more managers involved in decision-making, the real-time and interactive requirements of information transmission are more and more high, which puts higher demands on business intelligence. The development of the next generation business intelligence requires that it not only provide strategic decision for senior decision-makers, but also provide tactical decision and operational decision for other level users in the enterprise. The middle-level, grass-roots managers and front-line employees are often unable to query and analyze data in time before the computer because they are out of work. At the same time, various smartphones and tablets released in recent years have been able to store large amounts of data, provide rich human-computer interaction, and even display complex visual data, making it possible to migrate business intelligence analytics data to mobile devices. However, the current computing power and storage capacity of mobile devices is still a big gap compared with PC. Cloud computing, with its strong computing power and storage power, provides a guarantee of terminal intelligence by running a large number of computing and storage operations in a cheap and efficient cloud, with PCs, mobile terminals and thin terminals as the end of cloud computing. , the Business intelligence analysis is delivered to all levels of users in the enterprise by mobile terminals such as the smart mobile phones carried by users. Even customers and business partners, this will enable enterprises to face the problem of rapid response to improve customer service, so as to win the competitive advantage, the enterprise resources are fully and flexible application.

1. Cloud computing and mobile business intelligence

1.1 Mobile Business Intelligence

With the rapid development of mobile Internet, mobile business intelligence has become a new development direction of business intelligence. The business intelligence based on mobile Internet application can not only extend the past limited business intelligence application to all levels of employees, improve the utilization rate of business intelligence system, but also make it possible for business intelligence system to provide real-time access analysis. Compared with the traditional business intelligence, mobile business intelligence has the following characteristics: first, real-time access to business intelligence applications anytime, anywhere, employees no longer because of geographical location and access restrictions; second, the information initiative and personalized customization, enterprises can be based on the actual situation in advance for all types of employees customized timely push information, can also according to user's personalized customization for information push service; third, real-time monitoring of early warning, when the emergency occurs, the system will directly send the alarm information to the user's mobile devices, which allows employees to take timely measures to avoid loss.

1.2 Cloud computing and mobile business intelligence

Cloud computing is a kind of new technology architecture that has arisen in recent years, which distributes computing tasks on a large number of computer resource pools, so that all kinds of application systems can acquire computing ability, storage space and various software services according to need. Cloud computing is a low-cost, secure, scalable, and so on, it provides services to users through virtual machines, mirroring deployment execution, and the combination of cloud computing and business intelligence will further promote the application of mobile business intelligence. First, cloud computing can enhance the ability of business intelligence to handle massive amounts of data. The operation and storage ability is migrated to the cloud through the network, can complete the data Processing task which the PC cannot cope with, the user only can complete the various operation on the PC machine, such as PC, mobile device and so on. And users do not need to add high-performance mobile devices for business intelligence applications, as cloud computing maximizes resource sharing and eliminates time and space constraints. In addition, the mass data processing of traditional business intelligence is concentrated in data Warehouse, in recent years, the development of data Warehouse has been able to massively parallel processing of large numbers and efficient management of analytical data, while Hadoop, which is strongly supported by Map/reduce framework, is a distributed architecture based on cluster, regardless of mass data processing , cost, operation speed or stability are better than data warehouse. Second, cloud computing can enhance the timeliness of business intelligence. Cloud computing enables business intelligence to obtain transaction data in a shorter period of time, perform stronger data analysis functions, provide timely feedback in case of emergency, and give full play to the real-time advantage of mobile business intelligence. Third, the combination of cloud computing and business intelligence will reduce costs. Traditional business intelligence deployments based on data warehouses are expensive, and in addition to business intelligence and data warehousing, it costs and deployments, front-end business intelligence presentation tools, and the creation of large-scale data centers in low-cost areas can help businesses reduce costs in terms of power, network bandwidth, software, and hardware. Small and medium-sized businesses that have not been able to build their own business intelligence platforms in the past need not buy any hardware, need not install software, and simply hire mobile business intelligence applications in the cloud via SaaS mode. The combination of cloud computing and business intelligence enables companies to get flexible and resilient business intelligence services at less cost.

2. Mobile Business intelligence architecture based on cloud computing

The system uses the virtualization technology of cloud computing to build infrastructure layer, platform layer, support layer and application layer, and its architecture is shown in Figure 1.

Application layer provides a variety of business intelligence applications, such as data mining, OLAP, ad hoc query and real-time transaction processing. PC and thin client users can access and use these services through browsers or specific clients, and users of mobile devices can use all kinds of mobile terminals to use the service through mobile communication network or wireless network connection platform, and the system will provide different interface and function for users who hold different types of devices.

The support layer is responsible for handling the application request in the application layer and intelligently dispatching its business processing logic. The business intelligence system is confronted with huge and different requests for storage and computation, which can be divided into massive data processing request, daily information processing request, high performance computation request and so on. Intelligent Dispatch is judged by the type of request, if the request is very small, it can be found in the relational database to access the relational database directly, and the cloud computing platform is required to perform parallel computation if the request requires a large amount of computation and storage. Because the user's mobile device may be a tablet computer and other high-performance smart devices, may also be a more common low performance mobile phone, intelligent optimization of various mobile devices to identify different storage and computing capabilities and based on user request content measurement, for different types of mobile devices to provide the most suitable for their storage and computing capabilities of the browsing results.

(Responsible editor: The good of the Legacy)