[Twisted-Python] Benchmark of Python WSGI servers
itamar at itamarst.org
Sat Mar 19 10:26:47 EDT 2011
On Fri, 2011-03-18 at 20:54 -0400, Glyph Lefkowitz wrote:
> My pet theory is that it has something to do with transferring data
> from threads to the I/O loop via queue synchronization, and not being
> as smart as it could be about buffering, and that particular technique
> getting slammed really hard for very small request/response pairs. I
> hypothesize that more buffering would occur with larger responses with
> more chunks, and that would bring Twisted's performance up to those of
> these other servers.
Increasing the thread pool size to 50 gives a very nice speed boost (as
high as 100% boost in reqs/sec.) I emailed the benchmark author asking
if he would rerun with epoll and higher thread pool size.
More information about the Twisted-Python