A Comprehensive Survey on Fog Computing: State-of-the-art and Research Challenges

Cloud computing with three key aspects (ie IaaS, PaaS, and SaaS) and its inherent advantages (eg resilience and scalability) are still facing several challenges. For applications that are sensitive to latency, such as disaster management and content delivery applications, the distance between the cloud and the end device can be a problem. A service level agreement (SLA) may be processed even where the cloud provider does not have a data center. Calculating fog is a new example to solve such problems. Supports resources and services outside of the cloud, at the edge of the network, close to the end device, or finally specified by the SLA. Calculating fog is not a substitute for cloud computing, it is a powerful complement. Still supports edge processing while providing the possibility of interacting with the cloud. This paper presents a comprehensive survey of fog calculation. Critically review existing technology based on a concise series of evaluation criteria. Describes the architecture and algorithms for constructing a fog system. We will also introduce the issues and direction of research. In addition, the lessons learned will be reviewed and discussions on important roles that the fog may play in emerging technologies such as tactile Internet will be done.

The company is a developer of a global distributed fog computing platform. Basically, this system is a combination of many state-of-the-art technologies and open source code. SONM is based on an encryption protocol that provides assignment of tasks and validation of the results and pay for the computing capacity of the system used proportionately. Dbrain is an open block chain platform that can work in conjunction to build AI applications. Dbrain combines a lot of labor and data scientists, turning raw data into a real AI solution for business, creating a situation favorable for all participants in the AI ​​market. After receiving $ 2.5 million seed funds, Dbrain has reached an important milestone by releasing an alpha version of the platform including the web application and the Telegram bot.

In recent years, media promotes state-of-the-art technology, but the academic authority of parallel computing and computer microarchitecture shows the challenge of artificial intelligence - the limits of computing hinder the most challenging problems in the industry, AI The realization of the possibility. Xuran Peng, president of UIUC CSSA, says: "UIUC has a reputation for entrepreneurial spirit and innovation such as Youtube and Yelp." Students in China also enjoyed great success through a series of entrepreneurial examples such as Hesai Tech, Jisuanke, Kujiale. .... We believe it is a good idea to combine these benefits so that Chinese students can learn more about industry insights and professors from investors. "

Cloud computing with three key aspects (ie IaaS, PaaS, and SaaS) and its inherent advantages (eg resilience and scalability) are still facing several challenges. For applications that are sensitive to latency, such as disaster management and content delivery applications, the distance between the cloud and the end device can be a problem. A service level agreement (SLA) may be processed even where the cloud provider does not have a data center. Calculating fog is a new example to solve such problems. Supports resources and services outside of the cloud, at the edge of the network, close to the end device, or finally specified by the SLA. Calculating fog is not a substitute for cloud computing, it is a powerful complement. Still supports edge processing while providing the possibility of interacting with the cloud. This paper presents a comprehensive survey of fog calculation. Describes the architecture and algorithms for constructing a fog system.