Re:Application Load Balancers enables gRPC workloads with en...
by ComputerGuru on Friday October 30, 2020 @03:32PM
(#24945988)
> Really illustrates the dumbassery of sticking a (relatively) fast-moving application-layer protocol into the kernel.
Really? There are lots of problems with doing HTTP in the kernel, but “fast moving” is a new one. HTTP was stuck in permafrost from 1995 to 2015. If it weren’t for Google, neither of HTTP 2 or 3 would have ever happened.
Really? There are lots of problems with doing HTTP in the kernel, but “fast moving” is a new one. HTTP was stuck in permafrost from 1995 to 2015. If it weren’t for Google, neither of HTTP 2 or 3 would have ever happened.