The hottest AI chip edgetpu released by Google can

  • Detail

Google released AI chip edge TPU, which can be used to support IOT devices

[Abstract] this chip can be used in a variety of scenarios, but one of its initial uses is in the industrial manufacturing field: the consumer electronics manufacturer LG is testing this chip in a system that can detect manufacturing defects in the glass used for screens

Figure 1: during the i/o of the 2016 Google developer conference, its CEO, Sundar Pichai, was giving a speech. According to Tencent technology news on July 26, according to foreign media reports, Google is no longer satisfied with developing artificial intelligence (AI) chips for its own data center. It is now designing to integrate AI chips into products produced by other companies

two years after the release of tensor processing unit (TPU), Google launched edge TPU on Wednesday local time in the United States, which will enable sensors and other devices to process data faster

this chip can be used in a variety of scenarios, but one of its initial uses is in industrial manufacturing: the consumer electronics manufacturer LG is testing this chip in a system that can detect manufacturing defects in the glass used for screens

Google's entry into the "custom chip" market is a way for it to try to expand the market share of cloud computing and strengthen its competition with Amazon and Microsoft. Since 2015, Google has been using TPU to accelerate some workloads in its data center, rather than relying on commercial hardware provided by suppliers such as NVIDIA

2017, Google said that its AI chip was becoming more strategic. In the AI field, researchers are using a large amount of data to train models, so that the machine can carry out Goldilocks automotive materials when new data arrives, and constantly launch new technologies and products that conform to the development trend of automotive materials to pass our lens test

The original version of the

tpu can only make these predictions, while the second version (released in 2017) can be used to train the model. This update enables it to compete with the functional torque tester of nvidi Jinan new era Testing Instrument Co., Ltd. as a precision instrument a graphics card. The third generation TPU was released earlier this year

now we have the edge TPU, which is a microchip specially used to process the AI prediction part. It has less computing power than the training model. Edge TPU can run calculations by itself without connecting to multiple powerful computers, so applications can work faster and more reliably. They can handle AI work together with standard chips or microcontrollers in sensors or switching devices

joining Rhee, former chief technology officer of Samsung, said that Google did not let edge TPU compete with traditional chips, which was very beneficial to all silicon chip suppliers and equipment manufacturers. Edge TPU may "subvert the cloud computing competition" because many computing can now be performed on devices rather than all sent to the data center. In terms of cost and energy consumption, Google chips are more efficient than traditional chips in some types of computing

Google is not the only cloud computing service provider interested in the so-called IOT. The core of IOT is to manage and process data from many small embedded devices. Earlier this year, Microsoft announced the design of its IOT chip. Google's new chip will run a model based on a simplified version of tensorflow AI software, which the company released through an open source license in 2015

lg the CNS team responsible for helping internal and other companies handle it services is already testing edge TPUs and plans to start using them to inspect equipment on internal production lines

at present, in the process of producing glass for display panels, the detection equipment can process more than 200 glass images per second. Hyun shingyoon, chief technology officer of LG's CNS team, said that any problem that arises needs to be checked manually. The accuracy rate of the existing system is about 50%. The accuracy of Google AI can reach 99.9%

hyun shingyoon also said, "my expectation is to save money in discovering anomalies and defects that really affect our quality." His team had previously studied a computing system at NVIDIA

Google has built a toolkit, including edge TPU, NXP chip and Wi Fi connection, for developers to try. The company is working with manufacturers such as arm, HARTING, Hitachi vantara, nexcom, Nokia and NXP

joining Rhee did not disclose whether Google plans to build a more powerful edge TPU for the training model. (compile/Jinlu)

Copyright © 2011 JIN SHI