Industry News
News
Home » News » Details
News
Cloud computing is the basis of IOT and networking guide Cloud Architecture

China Software: all the potential advantage to full play of the Internet of things, must be based on cloud computing as the Internet of things. Hidden in the Internet was the philosophy behind the data collected most should be online transmission, only in this way can effectively gather, analyze the application and use of these data.

        With the development of the times, more and more of what the concept of the Internet of things. In addition to the built-in internal sensor and processor, these things are linked directly to the network, transmit their data. Although home automation is the concept of "main" play, such as the refrigerator milk without, the refrigerator will automatically from the grocery store where ordered milk. But the application of the Internet of things actually are growing. We will have to interact with each other and yet separate office things, are in need when automatic ordering office supplies, without our intervention, or even our clothes and body sensor on the real-time health data we transfer our doctor. This type of M2M (machine to machine) communication is the key.

        All the potential advantage to full play of the Internet of things, must be based on cloud computing as the Internet of things. Hidden in the Internet was the philosophy behind the data collected most should be online transmission, only in this way can effectively gather, analyze the application and use of these data. Now let's look back at the refrigerator example. In this example, not their refrigerator from the grocery store where ordered milk, but the refrigerator to transmit all the data of them, including the current consumption of food storage and the user, and then by the application to read and analyze these data. Then, considering the other factors, such as the user's current food budget fund and milk in the long delivery time and other factors, and then decide whether to buy, and the cloud is an ideal destination for these applications.

        If our daily products installed in all this, so the amount of data generated will be very large. Therefore, the Internet of things must consider how to store and analyze these generated data. This is not only an amount of data, which also involved the problem of the data generation rate. Sensors are generating more and more data, but these data generation rate has exceeded the speed of processing in most commercial applications.

        Cloud based solutions is the foundation for dealing with the data quantity and the speed of. The clouds can according to our requirements automatically to provide dynamic reserve storage resources, without manual intervention. The clouds also gives us through the cluster cloud database or the ability of physical storage virtualization can be adjusted without stopping the machine capacity access virtual storage and access large storage resource pool capacity, these are locally which can not be realized.

        Second questions about these data is how to deal with them. This problem has two difficulties. The first difficulty is how to real time processing all the data points obtained from each different objects there. A second problem is to extract useful information from all the available data points, and the information obtained from different objects there, add real value for stored data.

        Although real-time processing seems very simple -- to receive data, analysis of data, and then using these data -- but the real situation is not so. Let us look back at the refrigerator for example, imagine that every time someone opening the door of the refrigerator, the refrigerator is to send a packet, the data packet which includes the things moved, what is put in. We estimate, about 2 billion refrigerators worldwide, every switch refrigerator door 4 times, then the day will generate 8 billion packets per second, on average about 0.1 million packets, this volume is very alarming. What is worse, time characteristics of these data points may be mainly concentrated in one day (mainly is the morning and evening). If we according to the processing capacity of maximum load, so a lot of infrastructure will be wasted.

Once processed, then we will meet second problems, namely how to extract useful information from these data is stored, so that they are more the last step, and is no longer the personal affairs. If the refrigerator can automatically to your order to food store, for you personally this is very good, but if the manufacturers know the refrigerator from certain areas of overheating trend, or is the refrigerator storage some kinds of life consumption too fast, so the manufacturers said the value will be greater. In order to extract this information from the stored data, we need to use large data existing solutions (as well as some upcoming solutions).

        Cloud computing is very suitable to deal with these problems. The first difficulty, to allow for dynamic assignment (and recovery) processing resources, to require the application of real-time analysis of refrigerator data to deal with these data and to optimize the infrastructure cost. In the second difficulties, cloud computing can solve and large data plan to collaborate.

        To sum up, things may change the overall architecture of cloud computing, but at the same time the cloud computing to achieve this change is also very critical. A computing resource virtualization, although without manual intervention you can dynamically allocate these resources, but if so cloud computing won't have any development. Because of the Internet of things is the only motive force to promote their development.

Source: CIO times

Home      |      Business      |      News      |      Partners       |      Recruitment      |      About Us      |      Contact Us      |      Privacy
Copyright © 2011 Cetron Network Technology Co.,Ltd
:    鍏夊ぇ褰╃エ浠g悊   鐪熼挶妫嬬墝-瀹樼綉骞冲彴鎵嬫満app   缃戜笂鐪熼挶妫嬬墝--瓒呭ソ鐜╃殑鐜伴噾妫嬬墝娓告垙     鍏夊ぇ褰╃エ浠g悊