Research on data redundancy in cloud architecture

1 Introduction

With the development of science and technology, the environment complexity of IT industry is increasing, more http://www.aliyun.com/zixun/aggregation/13546.html ">it budget is used to maintain the infrastructure and maintain the normal operation of the application , rather than better meet customer needs. The emergence of the cloud seems to bring new hope to people, so some people make such a prediction-the future cloud era.

2 Cloud Architecture Design

2.1 Concept and characteristics of cloud architecture

The concept of cloud architecture, first proposed by Google in 2006, is a software architecture designed to meet on-demand services. Cloud architecture is different from the traditional computer system we are familiar with, the cloud architecture is a layer structure, the entire cloud architecture layer is divided into the display layer, the middle tier, the infrastructure layer and the management layer four levels. The display layer is primarily used to show the user what is needed, and to use the cloud Technology middleware layer of HTML, Javaseript, CSS, and Flash to provide more services to users. The middle layer plays a transitional role, it not only provides the support service for the display layer, but also provides the resources convenience for the infrastructure layer, and can also be used directly by the users. The middle tier is implemented primarily through 5 cloud technologies: REST, multi-tenant, parallel processing, application server, and distributed caching. Management is for the above three tiers of service, it provides account management, SLA monitoring, security management technology, to more comprehensively manage and maintain three horizontal layer of stability. Through the above analysis, the entire cloud architecture layer can be represented by the following framework (see Figure 1):

Figure 1 Cloud architecture Layer

The characteristics of the cloud architecture are based on the cloud architecture layer, which has the following characteristics:

(1) Large scale. "Cloud" system covers a large number of capacity, such as IBM, Microsoft, Apple and other "cloud" with thousands of servers running, "cloud" computing power and data capacity is not comparable to the general computer technology.

(2) strong data security. The management layer in the cloud architecture layer monitors and protects the data, account and other resources from unnecessary encroachment, and the management can respond to the 3-tier burst with its own service function, so the data stored in the cloud architecture is quite secure.

(3) Strong universality. Cloud architecture is supported by the cloud and can be controlled by unpredictable applications, and a "cloud" can control many different applications. (4) Strong scalability. Cloud architecture can change its optimal capacity and achieve optimal state at any time according to the needs of users.

2.2 Cloud Architecture Design

To achieve the powerful function of the cloud, the prerequisite is to have a complete cloud architecture. Cloud architecture is a computational schema that provides resources, data, and applications in the form of services to the client through the cloud. Based on previous research, cloud architecture mainly includes service control cluster, storage node cluster, compute node cluster, application node cluster, input and output devices. The structure of the cloud architecture layer is analyzed, so the cloud architecture layer can be expanded to specifically design the entire cloud architecture system (see Figure 2).

Figure 2 Cloud Architecture Design

Redundancy processing in 3 cloud architecture

Redundancy refers to the repetition of data, which is called redundant data. Data redundancy will not only affect the integrity of the database, but also waste storage system resources. Minimizing data redundancy is one of the main goals of cloud architecture design. Like the normalization theory of relational model, the main idea of computer system dealing with redundancy problem is the principle of minimum redundancy.

Cloud computing framework can better manage storage space, using the data redundancy processing mechanism of cloud computing, dividing data into several files, stored in different data storage devices of cloud, and if there is any abnormal storage device, it will not affect the integrity of storage data. The cloud uses node to extract file eigenvalues, reduce data duplication, and create data redundancy information in the cloud.

As shown in Figure 2, there are redundancy policies in the cloud architecture that indicate that there may be a large amount of data redundancy in the cloud architecture store. Because some of the associated data may need to be stored repeatedly, and reduce the amount of the same data, so that the association paradigm is optimized, is the future cloud architecture to take the road after all. The limitations of the associated data are reflected in the duplication of data storage, resulting in redundancy, increasing the computational volume of the cloud architecture, not only wasting storage space, but also slowing down the speed of the computer. The limitation of processing ability is shown in the search for simple information, the masking of complex information and the chaos of dynamic information. Because of the data redundancy, it makes the computer system running slowly and maintenance difficult, and can not meet the customer's demand.

Because cloud storage services adopt distributed data storage, redundant data is stored in a large number of storage devices in a clouds system to improve the system's destructive resistance. In the cloud architecture, there are mainly replica backup redundancy, for its backup can use the different storage files, reduce their storage, handle the data redundancy in the cloud architecture, save resources, reach the optimal storage system.

4 Conclusion

Handle the redundancy of the cloud architecture to get the most out of the least amount of input. The future will be the cloud of an era, the building and operation of cloud architecture gradually rise, it improves the efficiency of the calculation, minimize the duplication of data, and can effectively and timely processing of information, so that can be proactive.

(Responsible editor: The good of the Legacy)