Because sensors generate too much data and are difficult to pass to the cloud, edge computing is becoming a mainstream trend. Let's take a look at the relevant content.
According to Semiconductor Engineering, the original idea of ​​the Internet of Things (IoT) device was that a simple sensor would transfer raw data to the cloud and process it through one or more gateways. These gateways may be located in companies, homes, factories, and even connected vehicles. But it is increasingly obvious that there is too much data to be processed. This method is not feasible.
According to TIen Shiah, who is responsible for HBM marketing at Samsung Electronics, a PC will generate 90MB of data per day. One self-driving car produces 4 TB per day and the connected aircraft is 50 TB. Most of them are useless data.
Pre-processing, if done locally, requires less data in the cloud to achieve better performance at lower cost and less power, enabling the rapid response required for self-driving, drones and even robots. These are the reasons why edge computing suddenly gets so much attention. It brings the computing task closer to the data source, and in terms of self-driving, the final operation may be done in the sensor itself.
This is also important for artificial intelligence (AI), machine learning (ML), and deep learning (DL) applications. The key to AI/ML/DL is the ability to make inferences on local devices to improve security and performance. However, the larger problem with inference is memory throughput. Frank Ferro, senior director of product management at Rambus, said storage is once again a bottleneck. Many emerging applications, whether AI or ADAS, require higher memory bandwidth.
In addition, most of these applications are battery powered or have to survive within a highly limited power budget, and the difficulty of developing such devices is becoming more challenging.
One of the biggest problems with edge computing is that it is a transformational technology that is defined as it evolves. It is still virtually impossible to order dedicated edge computing products that support a specific combination of IoT devices, infrastructure, and computing requirements.
NVIDIA announced in late March that it will work with ARM (ARM) to integrate the NVIDIA Deep Learning accelerator architecture with ARM's Project Trillium machine learning platform, allowing chipmakers to easily add machine learning capabilities to IoT devices. Intel also introduced 14 new Xeon processors in February.
Both Intel and NVIDIA/ARM products can add more processing power to the endpoints, but neither product is ideal for passing data back to the cloud. Zeus Kerravala, principal analyst at ZK Research, said that NVIDIA's partnership with ARM and Intel's announced edge processors are the basic products designed for devices, gateways, etc. that require increased processing power.
The home IoT market may end up surpassing IIoT, but IIoT is setting the pace and agenda. According to Julian Watson, an analyst at IHS Markit, a market research organization, the demand for IoT gateways with edge computing capabilities is growing. Demand comes mainly from three specific areas: bridging for low-power nodes that are not directly connected to the network, such as Bluetooth-based (BLE) or Zigbee-based sensors; filtering traffic, determining what data should be processed at the edge and What data is sent to the cloud; manage the security of these edge devices.
IHS Markit executive director Michael Howard believes that IoT/edge gateways should at least do the following: 1. Reduce the amount of raw data from IoT devices by integrating duplicate data. 2. Convert the data to a format readable by the upstream application. 3. Have an upstream application that can determine which data will be available and which device. 4. Contains information on how to organize and optimize your data.
Howard said that if the gateway fails to refine the raw data into compact and useful data, it will only push time and bandwidth. Processing must be done where the data occurs, preferably more than once.
All major system suppliers are eager to enter the market, but the demand for gateways is growing. This problem is more complicated than collecting temperature data from several sensors. Especially in IIoT, the traditional SCADA and other automation systems in each vertical market are usually closed, proprietary, unfriendly to new communication technologies, and impossible to get rid of quickly.
ARM CEO Simon Segars said that there are so many next big things (Next Big Thing) that will happen, and it's hard to tell where to start. New communication protocols, whether new technologies such as 5G, LoRA, and NBIoT, require a lot of innovation in semiconductor devices. Currently AI is driving cloud chips. At the edge is the inference that is driving innovation in design.
Semi-Automatic Washing Machine Spin Motor
Semi-Automatic Washing Machine Spin Motor,Motor For Wash Machine,Al Wire Spin Motor,Semi Automatic Washing Machine Dryer Motor
WUJIANG JINLONG ELECTRIC APPLIANCE CO., LTD , https://www.jinlongmotor.com