Mailing-List: contact dev-help@tomcat.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Tomcat Developers List" <dev@tomcat.apache.org>
Received-SPF: pass (asf.osuosl.org: domain of dakota.jack@gmail.com designates
 64.233.162.192 as permitted sender)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
        s=beta; d=gmail.com;
        h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:references;
        b=FsExhTRklQxT4F4iB11cy/g/4f+aE9CrF+Dr4jvdvqksNl6PaVAy7v48PL090j5nEv4KVWb8x9z6jkHCmbHxk0Dk1lVciaIKKiDj2LbeP+jg7h/RDj3q3BHs5ury984stcr9//OROujsvH9/kx0sVYhTXu8sTADBR2HTQ3S+JUI=
Message-ID: <e03014300606222243j2104dd62ta7b12b4531b4f6d0@mail.gmail.com>
Date: Thu, 22 Jun 2006 22:43:18 -0700
From: "Dakota Jack" <dakota.jack@gmail.com>
To: "Tomcat Developers List" <dev@tomcat.apache.org>
Subject: Re: NIO vs BIO speed
In-Reply-To: <449B312A.3050806@hanik.com>
MIME-Version: 1.0
Content-Type: multipart/related;
	boundary="----=_Part_5555_32555234.1151041398376"
References: <4496A724.5060107@hanik.com> <449AC497.9080304@hanik.com>
	 <449AC79E.1060107@apache.org> <449ACB4F.3070701@ptc.com>
	 <449ACF73.5030209@hanik.com> <449AD393.3070101@apache.org>
	 <449B2261.9010001@hanik.com> <449B294E.7020401@hanik.com>
	 <449B2BC7.1010000@apache.org> <449B312A.3050806@hanik.com>

------=_Part_5555_32555234.1151041398376
Content-Type: multipart/alternative;
	boundary="----=_Part_5554_32476286.1151041398376"

------=_Part_5554_32476286.1151041398376
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

I am interested in this code.  Is there a way I can see the code?  Thanks.


On 6/22/06, Filip Hanik - Dev Lists <devlists@hanik.com> wrote:
>
>  Remy Maucherat wrote:
>
> Filip Hanik - Dev Lists wrote:
>
> Here is another test that I ran from a remote machine, setting
> maxThreads="25" and ab concurrency to 50 and keepalive on.
> In this case, NIO is a lot faster. Turn off keepalive on ab, and we get
> similar results to previous run, where BIO is a tad faster.
>
>
> Scaling the thread per connection model is done by increasing the amount
> of threads. This particular test demonstrates the obvious.
>
> yes, what is interesting though, is my NIO connector is not really true
> NIO, as it ties up a thread while polling for data. The true implementation
> would have not invoked that thread yet, but for that I would have had to
> rewritten the connector from scratch as I couldn't have taken advantage of
> code already written and tested.
> The reason it was done like this, is cause that way I could use almost all
> the code from the APR connector.
> So you could call it a semi-"thread-per-connection" model, yet handles
> better than the true thread per connection model.
> APR does the same thing.
>
>
>
> If I understand the results correctly, the results would be acceptable on
> Unix.
>
> yes, I think they are looking pretty good. And I am fairly confident in
> this new code, as most of it, is old tested APR code.
>
> Filip
>
> --
>
>
> Filip Hanik
>


-- 
"You can lead a horse to water but you cannot make it float on its back."
~Dakota Jack~

------=_Part_5554_32476286.1151041398376
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

I am interested in this code.&nbsp; Is there a way I can see the code?&nbsp; Thanks.<br>
<br>
<br><br><div><span class="gmail_quote">On 6/22/06, <b class="gmail_sendername">Filip Hanik - Dev Lists</b> &lt;<a href="mailto:devlists@hanik.com">devlists@hanik.com</a>&gt; wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div>


<div bgcolor="#ffffff" text="#000000"></div><div><span class="q">
Remy Maucherat wrote:
<blockquote cite="http://mid449B2BC7.1010000@apache.org" type="cite">Filip
Hanik - Dev Lists wrote:
  <br>
  <blockquote type="cite">Here is another test that I ran from a remote
machine, setting maxThreads=&quot;25&quot; and ab concurrency to 50 and keepalive
on.
    <br>
In this case, NIO is a lot faster. Turn off keepalive on ab, and we get
similar results to previous run, where BIO is a tad faster.
    <br>
  </blockquote>
  <br>
Scaling the thread per connection model is done by increasing the
amount of threads. This particular test demonstrates the obvious.
  <br>
</blockquote></span></div><div>
yes, what is interesting though, is my NIO connector is not really true
NIO, as it ties up a thread while polling for data. The true
implementation would have not invoked that thread yet, but for that I
would have had to rewritten the connector from scratch as I couldn't
have taken advantage of code already written and tested.<br>
The reason it was done like this, is cause that way I could use almost
all the code from the APR connector.<br>
So you could call it a semi-&quot;thread-per-connection&quot; model, yet handles
better than the true thread per connection model.<br>
APR does the same thing.</div><div><span class="q"><br>
<br>
<blockquote cite="http://mid449B2BC7.1010000@apache.org" type="cite"><br>
If I understand the results correctly, the results would be acceptable
on Unix.
  <br>
</blockquote></span></div><div>
yes, I think they are looking pretty good. And I am fairly confident in
this new code, as most of it, is old tested APR code.<br>
<br>
Filip<br>
<br>
<div>-- <br>
<br>
<br>
Filip Hanik<br>
<img src="cid:part1.06080605.05050906@hanik.com">
</div>
</div>


</div></blockquote></div><br><br clear="all"><br>-- <br>&quot;You can lead a horse to water but you cannot make it float on its back.&quot;<br>~Dakota Jack~<br>

------=_Part_5554_32476286.1151041398376--

------=_Part_5555_32555234.1151041398376--