Peng: Definition and characteristics of cloud computing

Cloud computing is the 3rd quarter of 2007, the emergence of a term. The definition and connotation of it are different, we can find at least 100 kinds of statements on the Internet, but there is no accepted definition. In this paper, the advantages of some of the conclusions, the definition of cloud computing and the characteristics of the discussion, I hope to be able to set the wisdom of people to find a more appropriate version. The requirement for this definition is to capture the nature of cloud computing in the most refined language, to cover today's popular typical cloud computing solutions (including Google Cloud computing, Amazon Cloud computing, Salesforce cloud computing, cloud security, etc.), but to differentiate other relevant concepts such as Grid computing, Parallel computation, etc.).

Figure Google's search trend reflects "Cloud Computing", a word that sprang up in the 4th quarter of 2007.

Definition: Cloud computing (Cloud Computing) is a new business computing model. It distributes computing tasks on a large pool of computer-made resources, enabling various application systems to acquire computational power, storage space and various software services as needed.

Description：

This resource pool is called the cloud. The cloud is a virtual computing resource that can be maintained and managed, typically for some large server clusters, including compute servers, storage servers, broadband resources, and so on. Cloud computing centralizes all computing resources and is automatically managed by software without human involvement. This allows application providers to be more focused on their business, in favor of innovation and lower costs, without having to worry about tedious details.

Called "Cloud", because in some ways it has the characteristics of the cloud in reality: Clouds are generally large, the size of the cloud can be dynamically scaled, its boundaries are blurred, the cloud is erratic in the air, you cannot and need not determine its exact location, but it does exist somewhere.

It's called "cloud," and because one of the originator of cloud computing, Amazon, which used to be known as Grid computing, took a new name, "Elastic Computing Cloud" (EC2), and achieved commercial success.

One analogy: It's like switching from an old single generator to a centralized power supply model. It means that computing power can also be used as a commodity circulation, like gas, water and electricity, easy to use, low-cost. The biggest difference is that it is transmitted over the Internet.

Cloud computing is the development of parallel Computing (Parallel Computing), distributed Computing (distributed Computing) and Grid computing (grid Computing), or the business implementation of these computer science concepts. Cloud computing is the result of a hybrid evolution of virtualization (Virtualization), Utility Computing (Utility Computing), IaaS (infrastructure as a service), PaaS (Platform-as-service), SaaS (software, services), and so on.

In general, cloud computing can be counted as a commercial evolutionary version of Grid computing. As early as 2002, Peng on the traditional grid computing ideas have no practical problems, put forward the concept of computational pool [1][2]: "It is scattered around the high-performance computer with high-speed network connectivity, with specially designed middleware software organically bonded together, The Web interface accepts the computing requests of scientists from all over the world and assigns them to the right node to run. The computing pool can greatly improve the service quality and utilization of resources, and avoid the inefficiency and complexity of dividing the application across nodes, and can meet the practical requirements under the current conditions. "If you change the" High-performance computer "into" Server cluster "and replace" science worker "with" business user ", it is very close to the current cloud computing.

Cloud computing has the following characteristics:

(1) Super scale. "Cloud" has a considerable scale, Google Cloud computing already has more than 1 million servers, Amazon, IBM, Microsoft, Yahoo, etc. "cloud" has hundreds of thousands of servers. Corporate private cloud typically has hundreds of thousands of servers. The cloud gives users unprecedented computing power.

(2) Virtualization. Cloud computing enables users to obtain application services in any location using a variety of terminals. The requested resource is from the cloud rather than the fixed physical entity. Applications run somewhere in the cloud, but in fact the user does not need to know or worry about where the application is running. Only need a laptop or a mobile phone, you can through the network services to achieve everything we need, even such as supercomputing tasks.

(3) High reliability. The cloud uses such measures as data duplication fault tolerance and computing node isomorphism to ensure high reliability of services, and using cloud computing is more reliable than using a local computer.

(4) universality. Cloud computing is not specific to the application, in the "cloud" under the support of the structure of the ever-changing applications, the same "cloud" can support the operation of different applications.

(5) High scalability. The scale of the cloud can be scaled dynamically to meet the needs of application and user-scale growth.

(6) on-demand service. "Cloud" is a huge pool of resources that you buy on demand; The cloud can be billed like running water, electricity and gas.

(7) Extremely inexpensive. Because the "cloud" of special fault Tolerant measures can use extremely inexpensive nodes to form the cloud, "Cloud's automated centralized management enables a large number of enterprises to not burden the increasingly high cost of data center management, the" cloud "of the universality of the resource utilization than the traditional system significantly increased, so users can fully enjoy the" cloud "low-cost advantage, It often takes hundreds of of dollars and a few days to complete a task that previously required tens of thousands of dollars and months to complete.

References:

[2] Peng Liu, Yao Shi, San-li Li, Computing pool--a simplified and practical Grid Model, computational Second Workshop on Grid and cooperative Computing (GCC 2003), Full-circle, Dec 7-10, 2003, published into lecture Notes in Computer science ( LNCS), vol. 3032, Heidelberg:springer-verlag, 2004. http://www.chinagrid.net/show.aspx?id=1915&cid=48

Source: China Cloud computing http://www.chinacloud.cn Author: Peng
Original link: http://www.chinacloud.cn/show.aspx?id=741&cid=17

(This article belongs to Freedocs, welcomed the full text reprint, or in the premise of not changing the original article part of the reference or even rewrite, but need to indicate the original source and author.) )