In short, cloud computing is actually an implementation of architectures and methods for large-scale computing through virtualization. In cloud computing, resources and functions are provided as services to users. For example, e-commerce sites such as amazon need to handle millions of requests and transactions daily, how to ensure processing power, how to ensure storage, and how to ensure that these can be done in a simple manner and with good performance?
Virtualization is the key. In fact, virtualization is not limited to this kind of server virtualization technology that VMs or Xen provide for running virtual machines. Familiar Java Virtual Machine, Hadoop Distributed File System, virtual memory are actually different types of virtualization technology. Representing resources in an abstract or logical way is virtualization. After all, the resources of a single server are limited. Through virtualization, the resources of different servers can be provided as a unified whole, so that users feel that they have a super-powerful server. To give an example, Hadoop is now in full swing is a superb example of computing power virtualization. Hadoop uses Map to decompose a large task into many small tasks. These small tasks are assigned to Hadoop service instances on different servers to calculate, calculate the intermediate results, and finally combine the result sets through the reduce method. He does not need to see how many servers in the Hadoop instance are clustered to perform computational tasks for the computing requester, feeling the superb computing power of that "awesomely incomparable" computer.
In fact, information technology has always been around three themes, that is, "computing", "storage" and "communications." Corresponding to these topics have a lot of cloud computing products: In terms of computing, there are Amazon EC2, Google App Engine, etc .; in the storage area, there are Amazon S3, mozy, etc.; in the messaging communications such as Amazon SQS. I especially recommend to Google App Engine to play, to experience the charm of cloud computing. Google App Engine currently supports both Java and Python programming languages, Web programming with GWT and Django respectively, as well as support for free accounts, enough to play as a small web site.
So in the end in China who need cloud computing? I think small-capitals need money because they can save equipment purchasing costs by using cloud computing. In data centers, energy consumption is a large part of data center costs (usually more than half), and cloud computing can effectively increase resources Utilization, reducing waste of power; large companies also need, because there are tens of thousands of servers within a company such as IBM, the same problem of resource utilization. In fact, the daily life of our ordinary people can not be separated from cloud computing. For example, more and more people use mobile phones to access various Internet services such as search and navigation. Cloud computing can ensure the quality of service and make us truly enjoy it .
Cottage "Mei Lanfang" in a word: cloud computing era!
Readers interested in cloud computing, really recommend a virtualization and cloud computing research team written by our IBM China Research Institute of virtualization and cloud computing, the book is our past 4 years of research knowledge and The crystallization of results, introduces the concepts of virtualization and cloud computing from a neutral technical perspective, the key technologies and industry dynamics. Welcome to criticize!