Microsoft unveils Project Brainwave for Real-Time AI
Codenamed Project Brainwave, Microsoft announced its new deep learning acceleration platform for real-time artificial intelligence (AI). To achieve the goal that will utilize a neural architecture and bring powerful data processing solutions to the market.
System is design for real time AI, which means the system processes requests as fast as it receives them, with ultra-low latency.
“Real-time AI is becoming increasingly important as cloud infrastructures process live data streams, whether they be search queries, videos, sensor streams, or interactions with users.
Project Brainwave achieves unprecedented levels of demonstrated real-time AI performance on extremely challenging models.” – Said by Microsoft Engineer, Doug Burger.
The Project Brainwave system is built with three main layers:
- High-performance, distributed system architecture;
- Hardware DNN engine synthesized onto FPGAs; and
- Compiler and Run-time for low-friction deployment of trained models.
Project Brainwave leverages the massive FPGA infrastructure that Microsoft has been deploying over the past few years. FPGA chips offer reduce latency, and are more flexible compared to conventional CPUs, because they can be reprogrammed to execute varied task, and thus reduce the need of new hardware.
Microsoft plans to bring this powerful, real-time AI system to its cloud computing platform Azure, to provide the benefit from Project Brainwave directly to its customers, complementing the indirect access though services such as Bing.
For more update don't forget to Like and Share. :-)
Like Our Facebook Page - www.fb.com/RubanTheGeek
Follow Us on Twitter - www.twitter.com/ruban_thegeek
Follow Us on Twitter - www.twitter.com/ruban_thegeek
Comments
Post a Comment