Analysis of the present situation and future of cloud platform in China from the origin of cloud

The term cloud comes from August 9, 2006, when Google CEO Eric Schmidt first presented the concept of "cloud computing" (Cloud Computing) at a search engine conference. Cloud computing is not yet a recognized definition. National Institute of Standards and Technology NIST's definition of cloud computing: cloud Computing is a pay-as-you-go model that provides available, convenient, on-demand network access, shared pools with configurable computing resources, shared pools of resources including networks, servers, storage, applications, services, etc. These resources can be quickly delivered to customers via the cloud computing platform, with little or no management effort or minimal interaction with service providers. This definition is relatively appropriate, as the cloud computing technology continues to evolve, our understanding of cloud computing is also deepening, the future of the definition of cloud computing will be more stringent.

Cloud computing is distributed computing, parallel computing, Grid computing, multi-core computing, networked storage, virtualization, load balancing, and other traditional computer technology development to a certain stage, and the development of Internet technology integration products. The aim is to integrate countless nodes (i.e. computational entities) into a "giant" system with powerful computing power via the Internet, with the help of IaaS (infrastructure as a service, infrastructure as services), PaaS (Platform as a) , platform as a service), SaaS (Software as a service, software, services), and other business models, to provide powerful computing power to end users. Cloud computing can improve the processing capacity of cloud computing platform, reduce the processing burden of user terminals, so that user terminals can be simplified into a low-cost computing terminal, allowing users to enjoy the use of cloud computing platform on demand powerful computing processing capacity.

Cloud computing, known as the fourth it Industrial revolution after mainframe computers, PCs, and the Internet, has not only changed the mode of network application, but also will be the foundation for the strong growth of it, Internet of things, E-commerce and many other industries. It is expected that the next ten years will be the era of cloud computing. More and more information systems and information terminals will rely on the cloud computing platform to exist, more and more research and development and business processes will rely on the cloud platform to support.

Although cloud computing is Google's first advocacy, the company that really makes cloud computing a big business is Amazon. As early as the concept of cloud computing in Google, in 2002, Amazon launched the Amazon Web service products, known as the AWS Business. Amazon's research and development and business applications in cloud computing have continued. By 2006, when Google first proposed the concept of cloud computing, Amazon discovered that cloud computing was in line with its own set of technical frameworks, and that it was launching Amazon's cloud computing. Subsequently, Amazon's revenue in the cloud business has been a year of high innovation, the industry leader. Amazon's service offerings in cloud computing have also been emulated by other cloud computing providers. Amazon offers a range of cloud services offerings:

1. Elastic Calculation Cloud EC2

2. Simple Storage Service S3

3. Simple database service SimpleDB

4. Simple Queue Service Sqs

5. MapReduce Service

6. Content Push Service CloudFront

7. AWS Import/Export

8. Relational database Services

Google is a veteran in the field of cloud computing. Why? There are two reasons: first, Google has the largest number of servers in the world. In order to solve these servers distributed computing, scheduling, task allocation, coordinated operation, monitoring and so on, Google in the distributed field of research investment is considerable. Second, the previous incarnation of Cloud Computing is distributed computing, which is the further development of distributed computing. Google's three most outstanding contributions to cloud computing are distributed File System GFS, MapReduce, and BigTable. Since Google unveiled the three detailed design papers, it has laid the foundations of cloud computing (and, of course, some technologies, such as distributed lock Chubby). However, Google did not launch its cloud-computing product Gae (Google App Engine) until 2008, and Gae is a relatively single cloud service product. Google did not officially launch its cloud-computing infrastructure service until 2012, Google Compute Engine, which includes services such as Google Cloud storage and Google BigQuery. Since then, Google's cloud computing services are relatively complete. Of course, many of Google's products are in their own SaaS mode, such as Google Docs, can be seen as cloud products.

This paper introduces the background of so many foreign cloud computing, because the domestic cloud computing service providers are learning, modeled on Amazon, Google and other companies of cloud computing services products. Next, let's take a look at the form and characteristics of cloud computing.

Second, the form of cloud computing

Cloud computing is mainly in the following ways:

(1) Software as service SaaS

SaaS is a mode of providing software through the Internet, software providers will deploy the application software on their own servers, customers can according to their actual needs, through the Internet to the manufacturers to order the required Web application services, according to the number of services ordered and the length of time to pay vendors, And through the Internet to obtain the services provided by manufacturers. SaaS allows users to stop buying software, instead of renting web-based software to a provider to manage business activities without maintaining the software, the service providers are solely responsible for managing and maintaining the software, and the software vendors provide Internet applications to their customers while also providing off-line operations and local data storage. Allows users to use their ordered software and services anytime, anywhere.

(2) Utility calculation

Utility Computing (Utility computing) is a service-providing model in which service providers provide computing resources and infrastructure management that customers need and charge according to an application, not just at the rate. The utility computing model on demand uses a variety of flexible and effective technologies that may change the way the entire IT industry operates.

(3) Platform is the service PAAs

PAAs is a variant of SaaS, which refers to the business base platform as a service that is submitted to the user in SaaS mode. Therefore, PAAs is also an application of the SaaS model. PAAs can provide a middleware platform for enterprise customization research and development, including database and application server. PAAs can increase the amount of resources that are utilized on the Web platform.

(4) Management service provider MSP

Management Services is one of the earliest forms of cloud computing, is an application software, commonly used in service monitoring, desktop management, virus scanning, anti-spam services and so on.

(5) Cloud network services

Cloud network services are closely related to SaaS, which means that network service providers not only provide sophisticated Web applications, but also provide APIs to help developers expand functionality over the network. Typical examples are Google Maps, Sina Weibo and so on.

(6) Business service Platform

The Business services platform is a hybrid of SaaS and MSP, providing a service collector that combines with users, providing an interactive platform between users and service providers.

(7) Network integration

Network integration is the integration of cloud computing infrastructure, the use of a common "cloud bus", the existing cloud computing companies to integrate, user-friendly comparison and selection of service providers, to provide users with complete and reliable service.

Iii. Characteristics of cloud computing

(1) Super scale

The cloud computing platform has a huge scale. The enterprise builds the private cloud generally is starts from dozens of servers, until hundreds of, thousands of sets. Companies like Amazon, Microsoft, IBM, and others have at least tens of thousands of servers. by June 2011, Jon Koomey estimated that Google has about 900,000 servers, based on Google's 36 data centers worldwide. It can be seen that the "cloud" built by countless servers gives users unmatched computing power.

(2) on-demand service

The cloud computing platform is a huge pool of resources to buy on demand. The resources of the cloud computing platform can be billed as the water, electricity, and gas used in our lives.

(3) High availability

By integrating massive storage and high-performance computing capabilities, the cloud computing platform delivers higher quality of service. The cloud computing platform can automatically detect the failure nodes and eliminate the failure nodes without affecting the normal operation of the system. Cloud computing platform does not target specific applications, under the support of "cloud" can construct the ever-changing application, the same "cloud" can support different application operation at the same time.

(4) High scalability

The scale of the cloud computing platform can be scaled dynamically. Because the application is running on the virtual platform, the fixed resource without reservation is locked, so that users can dynamically increase and decrease IT resources according to the application demand, can meet the needs of specific period, specific application and the change of user scale.

(5) High reliability

The cloud computing platform uses data redundancy and distributed storage to ensure the reliability of the data, the fault tolerance of data and the interchangeability of compute nodes can ensure the reliability of cloud storage. Any node has a physical failure, the cloud computing platform will automatically transfer tasks to other physical resources to continue to run.

(6) Virtualization

Virtualization technology can make the physical server virtual many less performance virtual machines, cloud computing platform can monitor and manage all the virtual machines in the cloud, as well as flexible scheduling and allocation of resources. The cloud computing platform provides services. Cloud computing enables users to obtain application services in any location using a variety of terminals. The requested resource is from the cloud. The implementation mechanism of cloud services is transparent to users, and users can obtain the required services by using terminals.

(7) High cost performance

Because cloud computing platforms can be built with relatively inexpensive server nodes and automated centralized management, businesses do not have to bear high data center management costs. The commonality of cloud computing platform makes the utilization of resources more significant than that of the traditional system; end users need only a very low terminal, they can experience the rich application of cloud computing.