FOG COMPUTING研究中的初级概念

meiseeny

2020-06-28

”Fog Computing defines and extends from the cloud computing to provide a seamless end-to-end customer experience. Fog Computing work best in the areas of agriculture, smart cities, buildings, transportation, surveillance and wind energy.”

雾计算的定义主要是云的拓展，为使用者提供一个与云无缝衔接的体验。雾计算只要在农业、智慧城市、智慧建筑、智能交通、监控和风力发电等领域。

Edge and Fog are also the same thing
Fog is a replacement also for cloud
A Fog is new name also for existing architectures
Fog nodes are also in constrained devices
The Fog computing applicable to wireless environments
Fog creates new silos and also eliminate some physical silos
- 注释
  - 边缘计算和雾计算指代的基本是同一件事
  - 雾计算也包含云计算
  - 雾计算涵盖很多已经存在的架构
  - 雾计算在无线通信中也适用
  - 雾计算在使用新设备的同时取代了很多旧的物理设备

云计算的主要限制（边缘计算之惯例，不一定准确！）

-Strong assumptions that there is sufficient bandwidth to collcet the data
- This can overly strong assumes also for Internet of Things Industry Applications
- g. Energy Utility 0.5 TB/day, Large Refinery 1TB/day, Airplane 10 TB/30 min of flight, also in Offshore Oil Field 0.75 TB/Week.
  - 　　云计算假设有足够的带宽来传输数据
    - 　　但是在物联网工业应用中数据量和数据量的增速很快
      - 　　举例来说，发电厂 0.5TB/天；大型炼油厂 1TB/天，飞机 10TB/半小时航程，海上油井 0.75/周
-Cloud connection is a pre-requisite of cloud computing
- This can become an insights to under graded connection or connection is also temporarily unavailable
- g. Driver Assistance Applications
  - 　　云计算必须与云相连接
    - 　　但是在一些场景下未建立连接或者链接会是不是中断
      - 　　比如无人驾驶
-Cloud computing analytics centralises-Defining the lower bound reaction time of the system
- Some IoT systems need to also be able to wait for the data to get to the cloud
  - 　　云计算的数据分析依赖云数据中心，这样带来了响应时间与带宽的权衡问题
    - 　　比如一些物联网系统必须要等待数据传输到云端进行处理。
-Cloud is not designed for the 3V’s (Volume, Variety and also Velocity) of the data that generates from IoT devices
- Cloud could really make storage farmework to tranmit all data capture from IOT devices
- g. Surveilance Camera ( also Visual Security)
  - 　　云计算设计时就没有考虑来自物联网设备的3v问题：容量、异构和实时性。
    - 　　云端很难胜任所有来自物联网设备采集数据的存储
      - 　　比如监控摄像机和视频安保心疼

Why Fog Computing?

Synergetic but not exclusive
Share and also store data efficiently
Take local decisions when fog devices communicate also in peer-to-peer
Provide solution to minimze latency, conserving network bandwidth, protecting sensitive and also reducing cost
Support dense geographical distribution and also mobility
- 　　雾计算为什么能！
  - 　　协同却不独占
  - 共享存储并且高效存储
  - 本地决策并且可以本地组我通讯
  - 提供全套解决方案来减少延迟、节省网络带宽、保护隐私并减少成本
  - 提供密集分布式支持和移动性支持

Usage of Fog Sites

Data Caching
Computation Offloading
Real Time Data Processing
- 　　雾计算节点的用途
  - 　　数据缓存
  - 计算卸载
  - 实时数据处理

Fog Computing Concepts

Local Data Processing
Cache Data Management
Dense Geographical also in Distribution
Local Resource Pooling
Load-Balancing
Local Device Management
Latency Reduction also for better QoS
Edge Node Analytics
- 　　雾计算核心
  - 　　本地数据处理
  - 缓存数据管理
  - 密集部署，分布式支持
  - 本地资源池化
  - 本地负载均衡
  - 本地设备管理
  - 减少延迟，优化QoS
  - 本地决策

Fog Computing Tech’s in the Future

Machine Learning
Artificial Intelligence
Fog-Edge Nodes also for Real-time Data Analysis
Cloud computing also for Data Storage
- 　　未来的雾计算技术
  - 　　机器学习
  - 人工智能
  - 雾/边缘计算节点的实时数据分析
  - 取代云计算，同时可以进行数据存储