Hi,

I just now happened to fall over this blog post from February 2011:

   http://devblog.seomoz.org/2011/02/high-performance-libcurl-tips/

The most striking thing I take from that is issue 6, where the author claims
that our DNS cache makes his performance degrade a lot. Did anyone notice
anything like that and/or have any further details/ideas/comments ?

The issues 1A and 2 seem to be bugs and I believe we have fixed them already.

Issue 4 sounds like a c-ares bug, although I've not had it reported and on
Linux I cannot quite see how it would happen (the post doesn't say what
platform they're using).

The rest seem to be just ideas and recommendations on how to use the libcurl
API. I'm not entirely sure about all of them, but then they're also a matter
of taste and design principles and not necessarily strictly objective truths.

-- 
  / daniel.haxx.se
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette:  http://curl.haxx.se/mail/etiquette.html