The business model and development bottleneck of cloud computing

Whether cloud computing can be popularized and applied in large scale depends crucially on the innovation of business model and the breakthrough of technology bottleneck. Innovative business model is one of the most distinguishing features of cloud computing from the traditional IT industry, that is, the information product itself is weakened and the service is upgraded to a new level. Compared with traditional information products, cloud computing addresses two major challenges: high costs, many companies often unable to afford the high cost of basic investment, through cloud computing, the enterprise can obtain the application of the large enough computing power, without the support of this computing capacity of the underlying it hardware and software to pay the corresponding original cost of investment The second is instant pay, when needed, like the purchase of services to pay this computing capacity, according to the flow of fees can be paid, do not worry about computing equipment and resources of the day-to-day maintenance costs and idle cost. At present, enterprises in the promotion of this business model operation does not lack the success of the case. Amazon (http://www.aliyun.com/zixun/aggregation/2467.html ">amazon") uses flexible Computing Clouds (EC2) and simple storage services (S3) to provide computing and storage services to the enterprise, through storage servers, Bandwidth, CPU resources and monthly fees to collect fees. Monthly fee and telephone monthly fee similar, storage server, bandwidth by capacity, CPU based on duration (hours) calculation of charge, at present, cloud computing has become one of the fastest growing Amazon business.

At present, there are still many bottlenecks in cloud computing at the technical level. such as Google, Microsoft, IBM and Sun Joint laboratory researchers have summed up cloud computing's "service availability, data locking, data confidentiality and auditing, transmission bottlenecks, performance unpredictability, scalable storage, flaws in large distributed systems, rapid scaling, reputation sharing, software licensing" 10 Major issue. Specifically, it can be summed up as: first, data security and privacy are hidden dangers. All kinds of resources are concentrated in the cloud, the confidential data of individual and enterprise will be kept and maintained by external personnel, so how to protect the user's information security and guide users to use it is a big challenge to develop cloud computing.

The second is the lack of unification of industry standards. The standards of cloud computing environment development services are just emerging, and applications developed using a manufacturer's programming model are not easy to migrate to another vendor's platform, such as Google's app Engine, which develops apps in a way that is quite different from the way the app is developed on Microsoft Azure. Third, the current data access and storage model is difficult to use widely. Whether Amazon's S3 and simple DB services, or Microsoft Azure's data Services, prefer a simple storage model based on binary objects or a simpler hierarchical model. While this brings significant flexibility, it adds a burden to the application logic to explain the relationship between different data elements, and many transactional applications that rely on relational database structures do not fit into this data storage model. Four is the network broadband bottleneck limited wide range of promotion. The contradiction is more prominent in China. Because whether it is based on the cloud services of individuals or enterprises to access and use the data through the Internet, so stable, high-speed broadband is the user can access the data at any time of the guarantee, will not delay the response due to network congestion. The regulation of relevant legal provisions is not perfect. There are still many gaps in the legal system of Internet in our country, such as the user's privacy problem, the supervision of the super computing ability of tens of thousands of servers under the distributed management and so on, in addition, in this new business model, how to protect the cloud computing provider's value-added services is also urgent to resolve.

(Responsible editor: Lu Guang)