Re: [boost] [http] Formal review of Boost.Http

2015-08-09 23:38 GMT+08:00 Vinícius dos Santos Oliveira < vini.ipsmaker@gmail.com>:

2015-08-07 15:41 GMT-03:00 Vinícius dos Santos Oliveira < vini.ipsmaker@gmail.com>:

...

2015-08-07 15:24 GMT-03:00 Glen Fernandes :

...

Do you have any preliminary benchmarks that you could share during the review period?

Perhaps to: - cpp-netlib - poco - casablanca - proxygen

I could provide a sloopy benchmark (just static hello world) during this weekend

Application: get the host header, get the request uri, concatenate both and send as reply.

Tested using weighttp -n 20000 -c 1000 -k http://localhost:8080/

https://gist.github.com/vinipsmaker/c2455df44e53bb305ca1

You can see that performance of Boost.Http versus Pion was roughly the same, but Boost.Http created no extra threads. If I used one extra thread, I think I could see a speedup. Also, notice that the pion example didn't log any output to the standard output, while the Boost.Http is printing everything that happens.

Why spawning a coroutine at L200 while you don't actually need it?