Microsoft and Intel build an artificial intelligence platform with a latency of less than 1 millisecond
At the moment, the tide of artificial intelligence has hit, and Microsoft has not missed it. In the wave of artificial intelligence, Microsoft can take the lead and bring subversive changes to Microsoft. It is reported that Microsoft teamed up with Intel to create an efficient artificial intelligence platform that can handle complex operations.
With the rise of artificial intelligence, many technology companies have been working in this field. After Google retired from the artificial intelligence platform, Microsoft teamed up with Intel to create a new artificial intelligence platform, which will be applied to real-time data analysis and prediction in the future.
Google launched the TPU at this year's Developer Conference, driving the development of a more efficient artificial intelligence platform. Microsoft is working with Intel to create a new artificial intelligence platform that provides 39.5 TFLOPS computing through Intel's StraTIx 10 FPGA chip. Performance, which can achieve performance speeds below 1 millisecond delay.
Microsoft's artificial intelligence platform, called Project Brainwave, is based primarily on Microsoft's Azuer cloud computing technology and combines Intel's StraTIx 10 FPGA chip to provide sufficient performance to enable the artificial intelligence platform system to handle more complex operations and more. For timely operational performance.
At present, Microsoft has not revealed when it is expected to use this artificial intelligence platform extensively. However, in the future, Microsoft will have more and more computing modes related to this platform. For example, automatic driving, automatic control, security monitoring, etc. require real-time data analysis and judgment. This will continue to increase the platform demand for greater calculus efficiency and faster response times.
At this year's Developer Conference, Google selected the second-generation TPU design to further enhance real-time calculus performance. NVIDIA also emphasized the creation of the TensorFlow intelligent learning system during this year's GTC 2017, and additional support for learning frameworks such as Caffe and CogniTIve Toolkit. Tesla V1000, which is also aimed at accelerating deep learning training and creating a more responsive artificial intelligence system.
Usb Switch And Socket,Usb Kvm Switch,Usb 3 Switch,Usb Socket
ZHEJIANG HUAYAN ELECTRIC CO.,LTD , https://www.huayanelectric.com