At the 2017 Build, Microsoft’s annual developers conference, Microsoft Azure CTO Mark Russinovich disclosed several major developments in Microsoft’s deployment of Intel FPGA.
The deployment of FPGA offer several advantages to Microsoft’s burgeoning cloud network.
- Flexibility and scale
- Super low latency networking
- Increase in networking speed – this is enabled by the low latency network and will help organizations process big data workloads.
Over the past few years, Microsoft has been deploying FPGAs in Azure servers, creating a more flexible environment for developers.
Microsoft’s FPGAs are already being used for Bing searches, deep neural network development, and software defined networking acceleration.
There’s a long history behind this new development from Intel. The process of getting FPGA onto Microsoft Azure began in 2011 with Project Catapult, an experiment that was launched to make AI initiatives more efficient. This was after realizing that multicore CPUs weren’t keeping up with the demands for faster computer processing.
Of course, Microsoft still relies on GPUs for hardcore computer processing, but FPGAs are becoming more relevant and popular in some of Microsoft’s key services. For instance, FPGA are used to power the Bing search engine – they are key components of the hardware behind image searching.
Luxe Electronics is an independent distributor of electronic components specializing in FPGA. We work with Intel, Xilinx, Microsemi and Lattice Semiconductor.
To learn more about what Luxe can do for your company, visit our website.