On Wednesday 27 March 2002 17:08, Daniel Stenberg wrote:
> On Wed, 27 Mar 2002, Jacky Lam wrote:
> > > I don't have any answers to why this happens. Before we continue on
> > > this, can you please post more detailed and specific, both test results
> > > and details on how your tests have been performed?
> >
> > The program I am developing is a web browser. I just mark the
> > performance by downloading a page serveral times and taking the average.
> > With other parts of browser remains the same, downloading a local(HK)
> > page requires 14s for libcurl version, but 30s for libwww version.
> > However, for www.cnn.com, the result is just reversed. Libwww seems
> > downloading much faster.
>
> You mean the entire CNN front page, including all the images etc? Can that
> be because of A) your lack of persistency due to the cleanup/init all the
> time and perhaps B) libcurl does not offer pipelining.
>
> I would say that A is a likely reason, while B wouldn't make such a huge
> difference.
>

        After rewriting the code to make use of persistent connection (and fix a bug
in the job scheduling between download agents), the download speed of our
browser is comparable to opera, netscape. So, LibCurl doesn't have any
performance problem. In fact, faster than single-threaded libwww. Thx.

Jacky
Received on 2002-04-04