A concept
Technology is developing so fast, you know, any kind of technology is not born out of thin air, to produce each item of information technology, new applications are increasingly high-performance, real-time, low-power, low-latency requirements , with the current information systems and their structure is insufficient in terms of computing, storage, transmission, and contradictory product. (More than a metaphor, right, now 5G era , but I use the PHS , but I want to enjoy 5G, how to do it, buy a 5G mobile phone ah! Ah ha ha ha ha ..)
Edge computing can be applied in different disciplines and fields, so he does not have a strict uniform concept.
We can understand him to perform calculations at the edge of a new computing model , the core of his philosophy is " calculated to be closer to the source of the data !! " On the Edge computing model, computing resources closer to the data source, the network edge equipment already has enough computing power to achieve local processing of data, and sends the results to the cloud (this we can be understood as: play a network game, we started on a number, and then, after the game board number, do not networks can play, and ah, this archive is local, that number is the cloud, drawing games: mobile games Jiu).
The above language may be understood by a biological experiment, which is known to everyone knee-jerk reflex test: when bouncing a small hammer struck the knee, the calf will be uncontrolled. (This analogy, we can hammer percussion as a data source , see the nervous system as edge computing , while the brain was cloud computing (Obviously the brain is not involved in this activity, or that the brain is involved only after incoming)), this non-conditioned reflex can accelerate human reaction speed, to avoid greater harm people . The future is all interconnected era, it is impossible to make cloud computing brain of each device, and edge computing is to let the device has its own brain! !
Resource edge of the network include: smart phones (to understand what this sentence: edge devices are changing to data from a single consumer-oriented role for the dual role of both data producers and data consumers, while the use of network edge devices have gradually mobile phone real-time pattern recognition data, perform predictive analysis or optimization, intelligent processing and other functions), computer, Wifi access points, routers, set-top boxes and other embedded devices.
II. Things, cloud computing
Things technical purposes in accordance with the agreement of the kind of communication and the Internet to connect, exchange information, in order to achieve intelligent identification, positioning, tracking, monitoring, and management of Internet resources, data information between the physical and the physical shared in real time, achieved with intelligent real-time data collection, transmission, processing, execution.
Background of the data networking and other applications in geographically dispersed, and on the response time and safety higher demand (for example: the appropriate time, for example, whether the fire fighting equipment within the building may be determined according to their fire safety For example, with smartphone security of personal information when the edge is calculated).
While cloud computing provides efficient computing platform for big data processing, but the current network bandwidth growth lags far behind the growth rate data (Haha, is not to understand why Thanos ....), while the complex network of networks environment for network delay is difficult to have a breakthrough improvement, therefore, the conventional cloud computing model will be difficult to efficiently support real-time applications based on Internet of things, but the problem is this: Bandwidth + delay ! ! So, ah, long-winded sentence is: linear growth of centralized cloud computing has been unable to match the explosive growth of massive edge data, a single computing resource cloud-based computing model can not meet the real-time big data processing, security, energy consumption other needs.
So, ah, you said earlier too general, said here, mention should also calculate the edge where the good news:
1. greatly ease the pressure data center and network bandwidth . (Massive growth in only a small amount of data is critical data, mostly for temporary data, no long-term preservation, edge computing can take advantage of this feature, processing a large number of temporary data at the network edge, and thereby reduce the pressure on network bandwidth data center).
2. enhance the responsiveness of the service . (You know, the network transmission speed is limited to the development of communication technology edge computing services provided near the user, to ensure close lower network latency, simple routing network also reduce the volatility of this we can understand silly into two people alive from the near point device does not like the Well, ha ha ha).
3. To protect the privacy of data . (Storage and use of key line edge computing infrastructure to provide data privacy, will restrict the operation of private data inside the firewall is an effective means to improve safety.)
III. Comparison of the edge of computing and cloud computing
二者相比，边缘计算并不是为了取代云计算，而是对云计算的补充和延伸，为移动计算、物联网等提供更好的计算平台，边缘计算模型需要云计算中心的强大计算能力和海量存储的支持，而云计算也同样需要边缘计算中边缘设备对海量数据及隐私数据的处理，从而满足实时性、隐私保护和降低能耗等需求。边缘计算的架构是“端设备--边缘--云”三层模型，三层都可以为应用提供资源与服务，应用可以选择最优的配置方案。
PS：
中期答辩后的一点小心思，因为自己的毕设涉及到智能工厂，但总觉得自己的工厂貌似并不智能，做了一个系统，到头来还是让人来控制，显得非常的迟缓，如果设备具有自己处理事情的能力是不是就智能了呢，于是，在不经意间发现了边缘计算这个概念，很棒哦。
每次学点新知识就会感叹自己知道的太少，就想一直学下去，不过呢，受挫的时候也会想，哎呀呀，快点毕业吧，大好青春都还没处个对象呢。。
以上读书笔记来自于《边缘计算》这本书，个人感觉蛮好的，目前看了四章了，剩下的读书笔记慢慢写。
还有一点就是，个人感觉，写读书笔记很有用，起码不会像在自己瞎看的时候那样心不在焉！！