您当前的位置是:  首页 > technology > Corporate communications > Abstracts >
当前位置:首页 > technology > Corporate communications > Abstracts > The explosive growth of IoT devices and cloud computing models are moving toward “fog computing”

The explosive growth of IoT devices and cloud computing models are moving toward “fog computing”

2018-05-15 11:02:10 Author: Source:CTI ForumComment:0  Click:


Since the first IoT device was introduced in 1990, the Internet of Things has developed a lot. This is a toaster that can be turned on and off on the Internet. Twenty-seven years later, connected devices have changed from novelties to indispensable parts of everyday life.
Recent estimates show that adults spend an average of more than 4 hours a day on smartphones, and only mobile phones are also devices equipped with IoT sensor data. At present, 81% of adults have smart phones. Imagine how much data we will receive when 81% of adults have smart cars and smart homes.
Today, most of the data on IoT devices is processed in the cloud, which means that data generated in all corners of the globe is sent centrally to a few computers in the data center. However, as the number of IoT devices is expected to soar to 20 billion by 2020, the volume and speed of data sent over the Internet poses a severe challenge to cloud computing methods.
Increasing device connectivity will force IoT manufacturers to shift the cloud computing model from the cloud computing model to a new model called "fog computing" in 2018.
The more data access, the obvious problem of cloud computing
The development of the Internet of Things and artificial intelligence will bring in hundreds of millions of dollars worth of data. Widely distributed sensors, smart terminals, etc., generate large amounts of data at all times. Although cloud computing has "infinite" pools of computing and storage resources, cloud data centers are often centralized and far away from terminal devices. When faced with a large number of widely distributed terminal devices and the massive data collected, cloud data centers cannot Avoid three major problems to avoid:
  • Network congestion, if a large number of IoT and artificial intelligence applications are deployed in the cloud, there will be massive amounts of raw data flowing into the core network uninterruptedly, causing congestion of the core network.
  • With high delays, long distances between terminal devices and cloud data centers will result in higher network delays, while applications with high real-time requirements will be difficult to satisfy.
  • Reliability cannot be guaranteed. For applications that require high reliability and security, the distance from the terminal to the cloud platform is long, and the communication path is long. Therefore, the risk is high, and the cost of backup in the cloud is also high.
Therefore, in order to meet the needs of applications such as Internet of things and artificial intelligence, Fog Computing has emerged as an extension of cloud computing. The fog calculation was first proposed by Cisco. It is a distributed computing model that acts as a middle layer between cloud data centers and IoT devices/sensors. It provides computing, networking, and storage devices that allow cloud-based services to leave Networked devices and sensors are closer.
The fog computing mainly uses devices in the edge network, which can be traditional network devices such as routers, switches, and gateways in the network, or can be specially deployed local servers. The resource capabilities of these devices are much smaller than a data center, but their large number can compensate for the lack of a single device resource.
In the Internet of Things, fog can filter and aggregate user messages, anonymously process user data to ensure privacy, initially process data for real-time decision making, provide temporary storage to enhance the user experience, and cloud can be responsible for large-volume operations or long-term storage tasks. Complementary with the advantages of fog computing. Through fog calculation, some data that does not need to be put on the cloud can be directly processed and stored at the edge of the network to improve the efficiency of data analysis and processing, reduce delay, reduce network transmission pressure, and improve security. The fog computing is rapidly applied by the Internet of Things and artificial intelligence due to its wide geographical distribution, large-scale sensor networks with a large number of network nodes, support for high mobility and real-time interaction, as well as diverse hardware and software devices and online cloud analysis. The enterprise has accepted and gained widespread application, for example, applications such as M2M, human-machine collaboration, smart grid, intelligent transportation, smart home, smart medical, and driverless.
Unlike edge computing, fog computing can extend cloud-based services such as IaaS, PaaS, SaaS to the edge of the network, while edge computing focuses more on the end device. Fog computing can perform edge calculations, but in addition to the edge network, fog calculations can also be extended to the core network, that is, the components of the edge and core networks can all be used as an infrastructure for fog computing.
"Cloud" and "fog" typical cases and application scenarios
Fusion cloud platform and fog computing can reduce the expenditures of traditional IT procurement, management, and operation and maintenance through the cloud and export IaaS, PaaS, SaaS as cloud services. On the other hand, fog computing can ensure real-time data collection at the edge. , Extraction and analysis speed, improve the deployment and management efficiency of network resources, help to improve the man-machine collaboration efficiency, and provide technical support for enterprise business innovation and service quality improvement. The following are typical cases and application scenarios of the four industries "cloud" and "fog."
Industry
Based on Pivotal Cloud Foundry, GE created the Predix IoT PaaS platform, which integrates Dell intelligent simulation technology to achieve "data twins." Based on cloud computing, GE realized the optimization of the aircraft engine production process. At the same time, based on the fog calculation, GE realized the "self-healing" during the flight of the aircraft.
As a PaaS platform for the Internet of Things, GE Predix also helps manufacturing companies convert big data, the Internet of Things, and artificial intelligence into smart manufacturing capabilities, enabling data innovation. The GE Predix platform, which integrates cloud computing and fog computing with “digital twins”, helps manufacturing companies achieve “virtual-real” design and production integration, and provides them with cloud computing services.
Agriculture
Chitale Dairy is a dairy plant. Based on Dell technology virtualization technology, Chitale Dairy implemented ERP cloud deployment. Based on fog calculation, they installed sensors for dairy cows to conduct near-real-time data collection, analysis, and processing to achieve refined operations and ensure the monitoring, management, and optimization of the entire dairy product production process. At the same time, Chitale Dairy realized the automated management of dairy production processes through the cloud-based dairy lifecycle management platform. Through the Internet of Things and big data analysis, the entire process monitoring of food intake, feeding, health, milk quality and output of each cow was monitored. Analyze and achieve refinement and automation of dairy production.
By linking the cloud's overall business management with fog-optimized inter-farm collaboration and milk source monitoring and management, we have improved the efficiency of dairy product lifecycle management while improving the efficiency of collaboration and collaboration and accelerating the pace of business innovation.
Service industry
TopGolf is a golf club. Through the adoption of Dell's virtualization and hyper-convergence technologies, golf digital high-end service output capabilities have been formed. They have broken the traditional golf business model by transforming into digital. Through the Internet of Things, the RFID chip is embedded in the golf ball to realize real-time monitoring of each shot, each player and event, and based on fog calculation, the path of each shot and ball is tracked and analyzed in real time, and real-time integration is realized. .
TopGolf's business model incorporates cloud computing and fog computing, enabling real-time data monitoring, interaction, and management across data centers, cloud, and edge applications to meet real-time event monitoring, on-field interaction, pre-game player score analysis, social media, The need for big data analytics such as personalized data management for members.
Transportation industry
In intelligent transportation, information can be collected by sensors, real-time data analysis and traffic deployment to improve public safety. Through fog calculation, a fog node in the intelligent traffic control system can share the collected traffic information to ease traffic jams and locate traffic accidents during peak hours, and can relieve traffic conditions in traffic congestion areas through remote control. Meanwhile, in each user's phone and public transportation, the fog-based application allows the user to share and download content through nearby users without a continuous network connection.
In addition, automated vehicle safety systems, on-road monitoring systems, and public transportation ticketing systems can all collect large amounts of information from sensors and video data. The aggregated data will be transmitted to the cloud, and data extraction and analysis will be performed according to the user's needs. The real-time analysis of the edge data will then be performed based on fog calculations so as to provide users with fast and accurate information to ensure the smoothness and safety of public transportation.
Future fog calculations will play a major role
From business operation mode to working life style, smart Internet of things technology is profoundly changing human society. To make the Internet of Things have ubiquitous intelligence, we must make full use of the distributed computing, storage, communication and control capabilities in the network environment, and effectively improve the production efficiency or user experience through resource sharing mechanisms and collaborative service architectures.
Currently, the research and standardization of fog computing technology has just started. The main technical challenges and research hotspots we face are: how to establish trust relationships among fog computing nodes, how to promote adequate sharing of resources among them, and how to achieve efficient communication and close cooperation among multiple levels of clouds, fog, and edges How to achieve fair and on-demand distribution of complex tasks between heterogeneous nodes.
It can be foreseen that with the continuous development and application of fog computing technology, smart Internet of Things will be more and more convenient, more and more realistically drawing on and mapping the organizational structure and decision-making mechanisms of human society so that it can use more natural and more familiar The way to provide everyone with accessible, ubiquitous intelligence services.
[Disclaimer] This article only represents the author's own opinion and has nothing to do with the CTI Forum. The CTI Forum maintains its neutrality in the presentation of statements and opinions, and does not provide any express or implied warranty for the accuracy, reliability or completeness of the contents contained therein. Readers are for reference only, and please bear full responsibility for yourself.

Topics