[ad_1]
The existing local and centralized cloud infrastructure cannot meet the massive computing needs of these powerful applications, which require low latency (or data transmission delay) to smoothly transmit and obtain real-time access to data. In order to reduce latency and bandwidth usage and control costs, computing power and processes must be closer to the physical location of the data. solution? Shift computing power to local infrastructure at the “edge” of the network instead of relying on remote data centers.
Up to 90% of industrial enterprises According to Frost & Sullivan’s prediction, edge computing technology will be used by 2022, and IDC’s recent report (Registration required) found that 40% of all organizations will invest in edge computing next year. Bike Xie, vice president of engineering at AI technology provider Kneron, said: “Edge computing is necessary to realize the next generation of industrial revolution.” He explained that the future of artificial intelligence and other automation technologies depends on the decentralized edge, whether it is through the integration of things. Networking and other devices connected to distributed network nodes are still realizing artificial intelligence-supporting chips that can automatically build algorithm models.
“Edge computing is a complement to the cloud,” Xie said. “Like the cloud, edge technology enables application manufacturers to acquire and apply the data-driven knowledge that will power smart factories and products.”
Manufacturing goes to the edge
Moving to the edge of computing is the result of tremendous changes in the manufacturing industry in the past two decades. Manufacturers, whether they are producing industrial products, electronic equipment, or consumer products, have slowly but steadily transitioned to improving the automation and self-monitoring of systems and processes to improve the efficiency of producing products, maintaining equipment, and optimizing every link in the supply chain. .
As manufacturers implement more sensor-based, automation-driven devices, they will also generate more data than ever before. But often, data sets from sensor-based devices to centralized systems can quickly become unwieldy, reducing the speed of automation and making real-time applications inoperable.
Xie said that edge computing gives manufacturers the flexibility to choose how to process data to eliminate time lag and reduce bandwidth usage, as well as which data can be destroyed immediately after processing. “If data transmission to the cloud becomes a bottleneck, manufacturers can quickly process data at the edge, and if latency and bandwidth are not an issue, manufacturers can move some data to the cloud.” He added that not only is it closer to using data Processing data locally can save bandwidth and reduce costs, and because it can be processed immediately, data is more secure.
IDC predicts that by 2023, more than 50% of the new enterprise IT infrastructure deployed will be at the edge rather than the enterprise data center, and less than 10% in 2020.
An example of the transition from cloud to edge is Paul Savill, senior vice president of product management and services at Lumen, a technology company that provides edge computing platforms. Lumen was recently installed in a new million-square-foot factory. Savill said that robotic systems from about 50 different manufacturers rely on edge computing because they require a 5 millisecond delay to accurately control the robot. The deployment provides a secure connection from the edge application to the robot manufacturer’s data center, where “they collect information in real time.”
Savill said, but for long-term storage of data and machine learning and analysis applications, all of this is stored in the public cloud. Other larger workloads are processed in big data centers with “strong computing power”, which can quickly process large amounts of data.
Savill said: “The chain from public cloud to edge computing to on-premises is very important.” “It enables customers to take advantage of the latest advanced technology, thereby saving money and improving efficiency.”
[ad_2]
Source link
