tyle=3D"">

In fact, I just re=
alised that 'ab' test is rather restrictive.... So here's a bit more of =
an extended test:

# ab -k -n 1000 -c 32
=

Apache + ExecCGI:

"white-space:pre-wrap;">Requests per second:    14.26 [=
#/sec] (mean)

e-wrap;">Time per request:       2244.181 [ms] (me=
an)

Time per =
request:       70.131 [ms] (mean, across all concurrent r=
equests)

Apache + mod_per=
l (ModPerl::PerlRegistry): 

e:pre-wrap;">Requests per s=
econd: 132.14 [#/sec] (mean)

le=3D"white-space:pre-wrap;">Time per request:     &nbs=
p; 242.175 [ms] (mean)

;">Time per request:       7.568 [ms] (mean, acros=
s all concurrent requests)

Inte=
restingly, without Keepalives, the story is much the same:

>

# ab -n 1000 -c 32

Apache +=
ExecCGI:

n>Requests per second:    14.15 [#/sec] (mean)

<=
span style=3D"white-space:pre-wrap;">Time per request:   &nb=
sp;   2260.875 [ms] (mean)

:pre-wrap;">Time per request:       70.652 [ms] (m=
ean, across all concurrent requests)

=

Apache + mod_perl (ModPerl::PerlRegistry): 

n style=3D"white-space:pre-wrap;">Requests per second:   &nb=
sp;154.48 [#/sec] (mean)

ap;">Time per request:       207.140 [ms] (mean)r>

Time per reque=
st:       6.473 [ms] (mean, across all concurrent request=
s)

Running some benchmarks across var=
ious parts of my site made me realise I also had some RewriteRules in th=
e apache config that still had H=3Dcgi-script - changed those to H=3Dper=
l-script and saw similar improvements:

>ExecCGI - Requests per sec=
ond:    11.84 [#/sec] (mean)

"white-space:pre-wrap;">mod_perl - Requests per second:   &n=
bsp;130.97 [#/sec] (mean)

That's quite some=
gains for a days work.

r=3D"auto">

round-color:rgb(255, 255, 255);">--

white-space:pre-wrap;background-color:rgb(255, 255, 255);">Steven Haigh

=F0=9F=93=A7 pace:pre-wrap;">netwiz-at-crc.id.aubackground-color:rgb(255, 255, 255);"> =F0=9F=92=BB tps://www.crc.id.au/" style=3D"white-space:pre-wrap;">https://www.crc.id=
.au

=

On Sun, Feb 7, 2021 at 23:58, Steven Haigh <netwiz-at-crc.id.=
au> wrote:

y" dir=3D"auto">

Interestingly, I did get things working with ModPer=
l::PerlRegistry.

What I couldn't find *anyw=
here* is that the data I was loading in Template Toolkit was included in=
the file in the __DATA__ area - which causes mod_perl to fall over!
=

The only way I managed to find this was the fo=
llowing error in the *system* /var/log/httpd/error_log (didn't show up i=
n the vhost error_log!):

re-wrap;">readline() on unopened filehandle DATA at /usr/lib64/pe=
rl5/vendor_perl/Template/Provider.pm line 638.

<=
/div>

Took me a LONG time to find a vague post that reading in lines=
from <DATA> kills mod_perl. Not sure why - but I stripped all the=
templates out and put them in a file instead and re-wrote that bit of c=
ode, and things started working.

I had to f=
ix a few lib path issues, but after getting my head around that, most th=
ings seem to work as before - however I don't notice much of an improvem=
ent in execution times, I do see this improvement using 'ab -n 100 -c32'=
:

">Apache + ExecCGI: Requests per second:    13.50 [#/se=
c] (mean)

Apa=
che + mod_perl: Requests per second:    59.81 [#/sec] (mean)r>

This is obviously a good thing.
iv>

I haven't gotten into the preload or DBI sharing =
yet - as that'll end up needing a bit of a rewrite of code to take advan=
tage of. I'd be open to suggestions here from those who have done it in =
the past to save me going down some dead ends :D

"qt-geary-signature" dir=3D"auto">

ap;background-color:rgb(255, 255, 255);">

le=3D"white-space:pre-wrap;background-color:rgb(255, 255, 255);">--n>

b(255, 255, 255);">Steven Haigh

=F0=9F=93=A7 pace:pre-wrap;">netwiz-at-crc.id.aubackground-color:rgb(255, 255, 255);"> =F0=9F=92=BB tps://www.crc.id.au/" style=3D"white-space:pre-wrap;">https://www.crc.id=
.au

=

On Sun, Feb 7, 2021 at 12:49, James Smith <js5-at-sanger.acuk=
> wrote:

ion1">

As welsey said =E2=80=93=
try Registry, that was the standard way of using mod_perl to cache perl=
in the server  =E2=80=93 but your problem might be due to the note=
in PerlRun=E2=80=A6

0/api/ModPerl/PerlRun.html#Description">https://perl.apache.org/docs/2.0=
/api/ModPerl/PerlRun.html#Description
r:black;background-color:white;">:12pt;">META: document that for now we don't chdir() into the script's d=
ir, because it affects the whole process under threads.  style=3D"">f;">k.html">ModPerl::PerlRunPrefork should
be used by those who run only under prefork MPM.le=3D"">
{tbh most people don=E2=80=99t use mod perl under threads a=
nyway as there isn=E2=80=99t really a gain from using them}

It =
suggests you use ModPerl/PerlRunPrefork =E2=80=93 as this does an additi=
onal step to cd to the script directory =E2=80=93 which might be your is=
sue=E2=80=A6.

 =
;

idth:initial;border-left-width:initial;border-right-style:none;border-bo=
ttom-style:none;border-left-style:none;border-right-color:initial;border=
-bottom-color:initial;border-left-color:initial;border-image-source:init=
ial;border-image-slice:initial;border-image-width:initial;border-image-o=
utset:initial;border-image-repeat:initial;border-top-width:1pt;border-to=
p-style:solid;border-top-color:rgb(225, 225, 225);padding-top:3pt;paddin=
g-right:0cm;padding-bottom:0cm;padding-left:0cm;">

al">From: Steven=
Haigh <netwiz-at-crc.id.au>
Sent: 07 February 2021 01:00=

To: modperl-at-perl.apache.org
Subject: Moving ExecC=
GI to mod_perl - performance and custom 'modules' [EXT]

=

 

<=
p class=3D"qt-MsoNormal">Hi all,

mal"> 

So for many year=
s I've been slack and writing perl scripts to do various things - but ne=
ver needed more than the normal apache +ExecCGI and Template Toolkit.>

 

ass=3D"qt-MsoNormal">One of my sites has become a bit more popular, so I=
'd like to spend a bit of time on performance. Currently, I'm seeing ~30=
0-400ms of what I believe to be execution time of the script loading, ru=
nning, and then blatting its output to STDOUT
and the browser can go do its thing. 

"qt-MsoNormal"> 

I beli=
eve most of the delay would be to do with loading perl, its modules etc =
etc

 

>

I know that the current trend would be to re-=
write the entire site in a more modern, daemon based solution - and I st=
arted down the Mojolicious path - but the amount of re-writing to save 1=
/3rd of a second seems to be excessive

MsoNormal"> 

Would I be=
correct in thinking that mod_perl would help in this case?

>

 

MsoNormal">I did try a basic test, but I have a 'use functions' in all m=
y scripts that loads a .pm with some global vars and a lot of common sub=
s - and for whatever reason (can't find anything on Google as to why), n=
one of the subs are recognised
in the main script when loaded via ModPerl::PerlRun.

=

 

mal">So throwing it out to the list - am I on the right track? wasting m=
y time? or just a simple mistake?

-signature">

 

"qt-MsoNormal">round-position-x:initial;background-position-y:initial;background-size:i=
nitial;background-attachment:initial;background-origin:initial;backgroun=
d-clip:initial;background-color:white;">--

"qt-MsoNormal">round-position-x:initial;background-position-y:initial;background-size:i=
nitial;background-attachment:initial;background-origin:initial;backgroun=
d-clip:initial;background-color:white;">Steven Haigh =3D"color:black;background-color:white;">ont-family:"Segoe UI Emoji", sans-serif;">=F0=9F=93=A7<=
/span>ition-x:initial;background-position-y:initial;background-size:initial;ba=
ckground-attachment:initial;background-origin:initial;background-clip:in=
itial;background-color:white;"> u">netwiz-at-crc.id.aual;background-position-x:initial;background-position-y:initial;backgroun=
d-size:initial;background-attachment:initial;background-origin:initial;b=
ackground-clip:initial;background-color:white;"> olor:black;background-color:white;">amily:"Segoe UI Emoji", sans-serif;">=F0=9F=92=BB>-x:initial;background-position-y:initial;background-size:initial;backgro=
und-attachment:initial;background-origin:initial;background-clip:initial=
;background-color:white;"> nt.com/v2/url?u=3Dhttps-3A__www.crc.id.au_&d=3DDwMFaQ&c=3DD7ByGj=
S34AllFgecYw0iC6Zq7qlm8uclZFI0SqQnqBo&r=3DoH2yp0ge1ecj4oDX0XM7vQ&=
;m=3DbosoTbkecbnrPukObNK-5Duc1p3JTllIM7_FHhBYKW4&s=3DvQDi0ezyEZDOz86=
GVraerPdT76UjN2in3UdPh8fglRM&e=3D">https://www.crc.id.au
[crc.id.au]

--=20
The Wellcome Sanger Institute is operated by Genome Research=20
Limited, a charity registered in England with number 1021457 and a=20
company registered in England with number 2742969, whose registered=20
office is 215 Euston Road, London, NW1 2BE.

>