No subject

SSL handshake has read 3125 bytes and written 293 bytes

Down from:
SSL handshake has read 4720 bytes and written 293 bytes

I'm using a temporary SSL cert from startssl.com so this purely academic
now, but I guess when it comes time to go to production I can make a more
informed decision when buying a cert.

-- 
Matthew Nuzum
newz2000 on freenode, skype, linkedin, identi.ca and twitter

"An investment in knowledge pays the best interest." -Benjamin Franklin

--0016e64c1bbce954df04943df310
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div class=3D"gmail_quote">On Thu, Nov 4, 2010 at 11:14 AM, Tom Pohl <span =
dir=3D"ltr">&lt;<a href=3D"mailto:tom at tcpconsulting.com">tom at tcpconsulting.=
com</a>&gt;</span> wrote:<br><blockquote class=3D"gmail_quote" style=3D"mar=
gin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div style=3D"word-wrap:break-word">Matt, you got me thinking this morning =
that I *just* read an article about this very topic and it was driving me n=
uts that I couldn&#39;t find it. It was a link posted by @ioerror on twitte=
r and I finally found it!<div>
<br></div><div><a href=3D"http://www.imperialviolet.org/2010/06/25/overcloc=
king-ssl.html" target=3D"_blank">http://www.imperialviolet.org/2010/06/25/o=
verclocking-ssl.html</a></div><div><br></div><div>Thanks!</div><div>-Tom</d=
iv>
<div><br></div></div></blockquote><div><br></div><div>Great article, there =
is a follow up that Chris on IRC sent at=A0<a href=3D"http://journal.paul.q=
uerna.org/articles/2010/07/10/overclocking-mod_ssl/">http://journal.paul.qu=
erna.org/articles/2010/07/10/overclocking-mod_ssl/</a></div>
<div><br></div><div>Between the two I have made some tweaks to my SSL confi=
guration that has lowered the response time for remote connections by a goo=
d bit:</div><div><br></div><div><div>Time per request: =A0 =A0 =A0 302.624 =
[ms] (mean)</div>
<div>Time per request: =A0 =A0 =A0 302.624 [ms] (mean, across all concurren=
t requests)</div></div><div><br></div><div>from:</div><div><div>Time per re=
quest: =A0 =A0 =A0 347.496 [ms] (mean)</div><div>Time per request: =A0 =A0 =
=A0 347.496 [ms] (mean, across all concurrent requests)</div>
</div><div><br></div><div>Because an extra certificate was getting sent cau=
sing the request size to exceed 4KB. The diff is about exactly one ping. It=
 had no affect on the time when doing the test on localhost.</div><div>
<br></div><div>From openssl:</div><div><div>SSL handshake has read 3125 byt=
es and written 293 bytes</div></div><div><br></div><div>Down from:</div><di=
v><div>SSL handshake has read 4720 bytes and written 293 bytes</div></div>
<div><br></div><div>I&#39;m using a temporary SSL cert from <a href=3D"http=
://startssl.com">startssl.com</a> so this purely academic now, but I guess =
when it comes time to go to production I can make a more informed decision =
when buying a cert.</div>
</div><br>-- <br>Matthew Nuzum<br>newz2000 on freenode, skype, linkedin, <a=
 href=3D"http://identi.ca" target=3D"_blank">identi.ca</a> and twitter<br><=
br>&quot;An investment in knowledge pays the best interest.&quot; -Benjamin=
 Franklin <br>
<br>

--0016e64c1bbce954df04943df310--