Hi guys,
On 5/25/07, peter royal <[EMAIL PROTECTED]> wrote:
On May 24, 2007, at 8:45 AM, Adam Fisk wrote:
> The slides were just posted from this Java One session claiming
> Grizzly
> blows MINA away performance-wise, and I'm just curious as to
> people's views
> on it. They present some interesting ideas about optimizing selector
> threading and ByteBuffer use.
>
> http://developers.sun.com/learning/javaoneonline/j1sessn.jsp?
> sessn=TS-2992&yr=2007&track=5
I just read over that this morning..
Without really knowing more details, my first blush is that on HTTP,
Grizzly is probably faster due to the fact that it was initially
tuned for HTTP. We know that MINA has extra latency wrt accepting
connections, so that's a definite spot that grizzly can beat us out at.
MINA was designed to be protocol agnostic from day one, whereas
Grizzly was designed for HTTP. (They have been working to remove the
HTTP-centric nature of the design from what I read on their blogs,
and their 1.5 release has easy support for doing any kind of protocol
now)
I didn't read the slide yet. Let me read it soon...
The HEAD revision of AsyncWeb highly depends on MINA and had
significant performance improvement. It seems like they used older
release (or didn't use a lightweight asyncweb example). Moreover, I
don't think MINA doesn't have any noticeably big margin for
improvement in performance, at least from the following performance
test report.
http://mina.apache.org/performance-test-reports.html
Trustin
--
what we call human nature is actually human habit
--
http://gleamynode.net/
--
PGP Key ID: 0x0255ECA6
