Fog computing ( English: Fog Computing ) or fog networking ( fogging ) , is the use of end-user end devices or edge devices connected to end-user devices for data storage in a distributed collaborative architecture (as opposed to centralizing data storage) in cloud data centers), or for distributed network packet transmission communications (as opposed to routing through the Internet backbone), or related distributed control or management. Fog computing is a concept proposed by Cisco (Cisco) in 2014 [1] . It is an extension of cloud computing. This architecture can process computing requirements hierarchically and regionally to resolve possible network traffic jams.
The application of fog computing is inseparable from the Internet of Things (IOT) and the Internet of Things (M2M). In the Internet of Things, most of the devices in our daily use will be connected to each other, and the most specific spiritual concept is to shorten the processing time of data, also TSN (Time Sensative Networking), most of the applications are medical, smart factories, smart cars, smart grids, etc. In this era of AI, fog computing can help background computing to operate more efficiently. It has the functions of computing, monitoring, and data collection close to the IOT DEVICE side, and then transmits the data to the data center for a large amount of data analysis. and calculus
Link : https://www.zhihu.com/question/50879568/answer/260088823
Source: Zhihu The
copyright belongs to the author. For commercial reprints, please contact the author for authorization, and for non-commercial reprints, please indicate the source.
Fog computing was originally started by Prof. Stolfo of Columbia University in New York, USA, and the intention at that time was to use "fog" to prevent hackers from invading.
Unexpectedly, Cisco of the United States grafted this term and carried it forward, which is the " fog computing" we are talking about now!
Fog computing: can be understood as localized cloud computing
The popular "cloud computing" is to put a large amount of data into the "cloud" for computing or storage, to solve problems such as insufficient computer or mobile phone storage, or insufficient computing speed
<img src="https://pic4.zhimg.com/50/v2-c5663cc3b0480382af3864de35b482c9_hd.jpg" data-size="normal" data-rawwidth="640" data-rawheight="409" class="origin_image zh-lightbox-thumb" width="640" data-original="https://pic4.zhimg.com/v2-c5663cc3b0480382af3864de35b482c9_r.jpg">
参自Cisco：雾计算原始定义图示
An additional layer is added between the terminal and the data center, called the network edge layer. For example, a small server or router with storage is added to directly process and store some data that does not need to be placed in the "cloud" at this layer to reduce the pressure on the "cloud", improve efficiency, and improve transmission. speed, reducing the delay, this working principle can actually be understood as: fog computing
Compared with cloud computing, fog computing is more grounded!
Cloud computing focuses on the way of computing, and fog computing emphasizes the location of computing.
Fog computing is closer to the ground than cloud computing! More specifically, they have different positions in the network topology!
There are actually many similarities between fog computing and cloud computing: for example, they are both based on virtualization technology, providing resource services for multiple users from a shared resource pool.
Compared with cloud computing, fog computing is closer to the place where data is generated. It is not unreasonable to say that " fog is closer to the ground than cloud" !
"Fog computing is between cloud computing and personal computing, and is a para-virtualized service computing architecture model. In addition, fog computing does not actually have strong computing power. Fog computing is to combine physically dispersed computers to form Weak computing power, but such computing power is completely sufficient for small and medium-sized data centers.”
If CDN is to make up for the problem of TCP/IP localized caching, then fog computing is to make up for the localized computing problem of cloud computing!
Fog computing VS cloud computing has the following obvious characteristics:
Lighter pressure: The computing resources are limited. Compared with the data center, the constituent unit of the cloud platform, the fog node is lighter! Fog computing can filter, such as aggregating user messages (such as continuously sent sensor messages), and only send necessary messages to the cloud, reducing core network pressure
Lower layer : Fog nodes are located lower in the network topology, have smaller network delays (total delay = network delay calculation delay), and are more reactive
More reliable: Fog nodes have a wide geographical distribution. In order to serve users in different regions, the same service will be deployed on fog nodes in each region, making high reliability an inherent attribute of fog computing. Once the service in a certain region is abnormal, User requests can quickly turn to other nearby areas to obtain related services. In addition, due to the use of fog computing, the amount of data sent to and from the cloud is reduced compared to cloud computing, and the delay is shorter than cloud computing, and security risks are further reduced!
Lower latency: In addition to the application of the Internet of Things, online games, video transmission, AR, etc. also require extremely low latency, which is also useful in fog computing.
More flexible: fog computing supports high mobility, mobile phones and other mobile devices can communicate directly with each other, and the signal does not have to go to the cloud or even the base station to go around! In addition, fog computing also supports real-time interaction, diverse hardware and software devices, and online cloud analysis.
More energy-saving : Fog computing nodes will not generate a lot of heat due to their geographical dispersion, so no additional cooling system is required, thus reducing power consumption, and fog computing is more power-efficient!
The emergence of fog computing, edge computing, sea computing, etc., is not used to replace cloud computing, but more to tinker with cloud computing "bug-like" problems, which are essentially born as an extension of cloud computing. products and concepts!