tensorflow-directml VS f-stack

and the fork of tensorflow is available here https://github.com/microsoft/tensorflow-directml

This might be outdated, but it was all I could find

Accelerating convolutional neural networks (ONNX and TensorFlow models) on GPU's (Nvidia/AMD/Intel/Qualcomm...). Since ML is pretty popular with dozens of frameworks out there all competing, it's not niche :b.

There is also https://github.com/microsoft/tensorflow-directml which should work on AMD. But I haven't used it since forever so I'm not sure of its current state.

So, we try to use Photon coroutine lib to simplify the development of DPDK applications with the new concurrency model, and provide more functionalities, such as lock, timer and file I/O. First of all, we need to choose a userspace network protocol stack. After investigation, we have chosen Tencent's open source F-Stack project, which has ported the entire FreeBSD 11.0 network protocol stack on top of DPDK. It also has made some code cuts, providing a set of POSIX APIs, such as socket, epoll, kqueue, etc. Of course, its epoll is also simulated by kqueue, since it is essentially FreeBSD.

I agree most HTTP server benchmarks are highly misleading in that way, and mention in my post how disappointed I am at the lack of good benchmarks. I also agree that typical HTTP servers would fall over at much lower new connection loads.

I'm talking about a hypothetical HTTPS server that used optimized kernel-bypass networking. Here's a kernel-bypass HTTP server benchmarked doing 50k new connections per core second while re-using nginx code: https://github.com/F-Stack/f-stack. But I don't know of anyone who's done something similar with HTTPS support.

Software for Internet Service Providers. The current project is based on DPDK, on top of it we use modified version of F-stack and then our application logic. There is some application logic "under" the F-stack too.